Increasing the Stability of Thylakoid Biocatalyst Bio-Solar Cells Through the Use of Oxygen Species Material Scavengers

Global energy requirements will only increase with time. While fossil fuels can be relied upon for several more centuries, they would produce vast amounts of carbonaceous byproducts. This undesirable fact makes renewable options like bio-solar cells, which are clean, inexpensive, and take advantage of abundant solar energy, a tempting prospect. However, bio-solar cells often have very short lifetimes due to reactive oxygen products that build up during photosynthesis. Past research using the oxygen scavenger catalase to reduce the quantity of these oxygen byproducts has had advantageous effects on bio-solar cell lifetimes, leading to interest in other types of oxygen scavengers. Two of these, ascorbic acid and activated carbon, were tested to compare their abilities at extending solar cell lifetimes with those of catalase. Amperometric and solar cell tests reveal that low concentrations of ascorbic acid are the most effective methods of extending bio-solar lifetime tested in this study, increasing the lifetime of bio-anodes to 148% of the lifetime of blank bio-anodes as compared to the 110% increase achieved by catalase. Additionally, it was determined that ascorbic acid bio-anodes produce high photocurrents with less uncertainty than either catalase or activated carbon conditions.

Increasing the Stability of Thylakoid Biocatalyst Bio-Solar Cells through the Use of Oxygen Species Material Scavengers By Megan Stephanz A design thesis submitted as a partial fulfillment of the requirements for the degree of Bachelor of Science in Materials Science and Engineering University of Utah April 26, 2016 Approved: ______________________________________ Shelley Minteer, Faculty Advisor, MSE ______________________________________ Ashutosh Tiwari, Senior Thesis Advisor, MSE ______________________________________ Dinesh K. Shetty, Honors Thesis Advisor, MSE ______________________________________ Russell W. Askren, CLEAR Instructor ______________________________________ Feng Liu, Department Chair, MSE ABSTRACT Global energy requirements will only increase with time. While fossil fuels can be relied upon for several more centuries, they would produce vast amounts of carbonaceous byproducts. This undesirable fact makes renewable options like bio-solar cells, which are clean, inexpensive, and take advantage of abundant solar energy, a tempting prospect. However, bio-solar cells often have very short lifetimes due to reactive oxygen products that build up during photosynthesis. Past research using the oxygen scavenger catalase to reduce the quantity of these oxygen byproducts has had advantageous effects on bio-solar cell lifetimes, leading to interest in other types of oxygen scavengers. Two of these, ascorbic acid and activated carbon, were tested to compare their abilities at extending solar cell lifetimes with those of catalase. Amperometric and solar cell tests reveal that low concentrations of ascorbic acid are the most effective methods of extending bio-solar lifetime tested in this study, increasing the lifetime of bio-anodes to 148% of the lifetime of blank bio-anodes as compared to the 110% increase achieved by catalase. Additionally, it was determined that ascorbic acid bio-anodes produce high photocurrents with less uncertainty than either catalase or activated carbon conditions. 2     1. INTRODUCTION Global energy demand is expected to grow at a prodigious rate over the next century. While fossil fuels can supply this demand for quite some time, they would do so at the cost of the planet's air quality and atmospheric carbon dioxide levels. Bio-solar cells are one possible solution to meet global energy demands without increasing air pollution and carbon dioxide concentrations. Made up of bio-anode sections coated with a biological, photosynthetic catalyst and connected to bio-cathodes, these photovoltaic devices use the electron transport generated by photosynthetic exposure to light to produce a photocurrent of electricity. The anodes are generally created by applying a concentrated solution of thylakoids (the photosynthetic membranes within the chloroplasts of plants and some bacteria) to a carbon paper electrode, then immobilizing the thylakoid membranes onto the surface by vapor depositing a thin overcoating of silica. The cathodes are generally created by applying solutions of reducing enzymes to new electrodes of carbon paper. All of the materials involved with creating these bio-solar cells are cheap (coming simply from plants or readily available silica), non-toxic, and easily disposable, giving them advantages over conventional semiconductor solar cells. However, several major problems are limiting their lifetimes. Without a level of longevity, these bio-solar cells will not be a viable energy solution. The largest problem limiting the lifetime of thylakoid bio-solar cells is the instability created by the products of photosynthesis. The photosystems within the thylakoids use sunlight to turn carbon dioxide and water into oxygen and carbohydrates, in the process creating the electron transport used by the solar cells to create a current. However, the isolated thylakoids have no method of removing the oxygen byproducts formed by photosynthesis, and, as a result, a 3     build-up of reactive oxygen species slows the photosynthetic reaction and damages the components of the thylakoids. The goal of this project is thus to design a method of reducing the harmful effects of reactive oxygen species accumulation on the lifetimes of thylakoid bio-solar cells. One proposed solution to this problem is the incorporation of oxygen scavenger species into the thylakoid anode. Material scavengers are chemical substances added to mixtures in order to remove or deactivate unwanted impurities or reaction products. The challenge, in this case, is to find an oxygen scavenger that can mix homogenously with the thylakoid membranes and effectively deactivate the reactive oxygen species without interfering with electron transport or device structure. 4     2. LITERATURE REVIEW 2.1 THE POTENTIAL OF SOLAR ENERGY By midcentury, global energy consumption is expected to double from beginning century levels due to population and economic growth [1]. While the currently available fossil fuel resources (especially coal reserves) could reasonably supply this demand for several centuries, consumption of fossil fuels at the rates required to do so would drastically increase the level of carbon dioxide in the Earth's atmosphere [1]. Thus, the development of renewable energy sources is becoming increasingly important. Particular interest is devoted to solar energy due to its great abundance: the amount of solar energy that strikes the Earth's surface in one hour is approximately equal to the global energy consumed in one year [2]. A number of photovoltaic devices have already been developed to use this abundant source of energy. Many of these photovoltaics utilize inorganic semiconductor materials. These devices, although they currently have the highest efficiency at converting solar energy to electricity (10-44%), often use expensive, frequently toxic materials [2]. Because it is highly unlikely that solar energy will become a popular substitute for fossil fuels if it continues to be much more costly, organic photovoltaics have also been considered. Unlike many semiconductor technologies, organic photovoltaics are non-toxic, disposable, and inexpensive [2]. In addition, biological organic photovoltaics take advantage of photosynthesis, nature's highly evolved method for solar energy conversion [2]. During photosynthesis, plants and certain types of bacteria use light from the sun to turn carbon dioxide and water into oxygen and carbohydrates, causing a cascade of electrons that provide an electric current in the process. As seen in Figure 1, photosynthetic reactions can 5     Figure 1: Depicting photosynthetic reactions as a thylakoid anode with a Calvin Cycle cathode [2] indeed be modeled as electrochemical reduction and oxidation reactions: within the chloroplasts, or photosynthetic organelles of plants, the thylakoid membranes act as an anode and oxidize water to form oxygen and energy carrying molecules like ATP and NADPH. These energy carrying molecules then allow the Calvin Cycle, or cathode of this model, to reduce carbon dioxide into a number of carbohydrates that the plant then uses as fuel. Within the thylakoids are systems of proteins called photosystems I and II that actually capture sunlight and convey the cascade of electrons. The quantum yields of photosystem I and photosystem II are close to 1 and 0.8 respectively, which is much higher than the less than 50% efficiency obtained by some of the more successful semiconductor solar cells [2]. In order to be of any use, however, there must be electron transfer between the biological catalyst and the electrode surface to which it is applied. Direct electron transfer (DET) is the ideal transfer method, as it moves electrons directly from the biocatalyst to the electrode without voltage loss or added instability. However, the active photosynthetic sites must be less than 1 6     nm away from the electrode surface to enable electron tunneling. If DET is not possible, a mediator chemical species must be added to the biocatalyst to carry electrons from the active sites to the electrode. This mediated electron transfer (MET) is often accompanied by voltage losses and instability [2]. Additionally, the creation of oxygen as one of the natural byproducts of photosynthesis leads to its own problems. When a high concentration of oxygen is trapped within a biological system, as is the case of a bio-electrode, the build up of product will slow down the chemical reaction itself. More importantly, the reactive oxygen species that are formed will damage the biological components performing photosynthesis, thus destabilizing the entire device [2]. While cheap, clean, and disposable, these bio-solar cells have yet to consistently achieve the 10% efficiency needed to make them serious competitors in the photovoltaic market in terms of lifetime value, and stability is an important issue that needs to be addressed, as inefficient microbe bio-solar cells have lasted only around one year and the more efficient thylakoid and photosystem bio-solar cells have lasted only 30 or 11 hours, respectively, depending on the dominant type of photosystem present [2]. Additionally, improving the DET photocurrent production is also crucial to future success of bio-solar cells. 2.2 CURRENT PROGRESS IN BIO-SOLAR CELLS In terms of past bio-solar cell research, the three most commonly investigated photobioelectrocatalysts (or bio-catalysts that use photosynthesis to enhance electricity production) are full, photosynthetic microbes, thylakoid membranes, and isolated photosystems. Full, photosynthetic microbes have all the biological hardware for complete photosynthesis and can replace any damaged cell components to increase stability. However, because of the many 7     membranes between the photosystems and electrode surface, MET is necessary. Thylakoid membranes, on the other hand, contain all the necessary components for photosynthesis but have no outer membranes, so DET is possible. Having said this, it is sometimes difficult to get their active sites close enough to the surface, and they have less stability due to their inability to repair themselves. Photosystems are even closer to the electrode surface and allow DET, but are unstable for similar reasons and more difficult to isolate than thylakoids [2]. Of the two biocatalysts which allow DET, thylakoids are the easiest to isolate from their source cells, so this biocatalyst was chosen for further study. Further research determined whether thylakoids from some sources were more efficient than others. In a study that compared the performance of bio-anodes made with thylakoids extracted from organic spinach, arugula, beets, green chard, kale, collard greens, and watercress, the spinach electrodes consistently had the highest photocurrents per chlorophyll concentration [3]. This is most likely due to the fact that the soft leaves of the spinach allowed for easy cell disruption to acquire the thylakoids without much exposure to cell vacuolar sap. This is important because fatty acids inside this sap could activate polyphenol oxidases, which would reduce photosynthetic activity. In addition to this, spinach also had a lower concentration of phenols and other inhibiting compounds than the other plants tested. This suggests that the best thylakoids for bio-solar applications come from sources with limited exposure to vacuolar sap and low concentrations of phenols and other inhibiting chemicals [3]. However, even with the most advantageous source plant used to provide thylakoids, the lifetimes of the bio-solar cells are still reduced by the reactive oxygen species produced during photosynthesis. In one experiment, catalase, an enzyme commonly employed to reduce the damaging effects due to reactive oxygen species production, was incorporated into thylakoid bio8     Figure 2: Schematic of a thylakoid bio-solar cell with catalase incorporated in the anode [4] anodes as shown in Figure 2 [4]. The test found that activity of bio-anodes without any additives was not stable and decreased to background value after less than 10 minutes. Bio-anodes made with the catalase oxygen scavenger, however, remained constant for approximately 30 minutes before gradually decreasing to 50% activity after 2 hours [4], indicating the success of the catalase at reducing harmful reactive oxygen species. Further research into methods of increasing thylakoid bio-anode efficiency found that utilizing stroma thylakoids more than grana thylakoids increases DET with the electrode [5], as does use of carbon quantum dots [6]. As a method of increasing surface area, thylakoid membranes in chloroplasts exist in a highly folded state. The majority of these folds consist of stacked discs, called grana thylakoids, connected by small membrane layers, called stroma thylakoids, as seen in Figure 3 [5]. The Figure 3: Depiction of grana vs. stroma thylakoids [5] 9     main reason stroma thylakoids are more effective at DET than grana is that their thinner nature allows more of the photosystem I on their surfaces to be close to the electrode surface than the highly stacked and folded grana [5]. In addition to increasing surface area, which allowed more DET between thylakoids and the electrode surface, carbon quantum dots can also fluoresce, allowing more of the solar spectrum encountered by electrodes to be used for solar energy conversion [6]. 2.3 PROBLEM AND DIRECTION As several methods are available for improving the efficiencies of these bio-solar cells, the most significant hindrance to their widespread use is the short lifetimes caused by reactive oxygen species. Because the oxygen scavenger catalase was able to make small improvements over device lifetime [4], this project focused on the design of a reactive oxygen species scavenger system that could increase device lifetime beyond catalase. This reactive oxygen species scavenger system must be biocompatible with thylakoids and have minimal electrochemical background. 2.4 OXYGEN SCAVENGERS AND THEIR METHODS Oxygen scavengers are chemicals or chemical systems that reduce the levels of oxygen and oxygen byproducts within an environment through a variety of methods. Much of oxygen scavenger research is motivated by the food and packaging industry due to oxygen scavengers' ability to maintain food product quality by decreasing food metabolism, reducing oxidative rancidity, inhibiting undesirable oxidation of pigments and vitamins, controlling enzymatic discoloration, and inhibiting growth of aerobic microorganisms [7]. Within this research, several varieties of oxygen scavengers have been investigated, such as metal or metal/halide systems and 10     various non-metallic systems including enzymatic methods, ascorbic acid, and activated carbon [7] [8]. The most commonly used metallic systems involve iron or sulfites. The majority of ferrous oxygen scavenger systems are based on the principle of iron oxidation in the presence of water. Sodium chloride and other salts are often added as catalysts to lower the humidity at which this reaction can occur [7]. Moisture permeates into iron particles, which activate and oxidize into iron oxide, as seen in the following reactions [8]. Fe à Fe2+ + 2e0.5 O2 + H2O + 2e- à 2 OHFe2+ + 2 OH- à Fe(OH)2 Fe(OH)2 + 0.25 O2 + 0.5 H2O à Fe(OH)3 In terms of this method's use in a biological system, iron, while essential to plant growth, is highly reactive and toxic via the Fenton reaction [9], during which iron reacts with hydrogen peroxide to form hydroxyl radicals, another reactive oxygen species [10]. Additionally, sodium is an element that can cause severe dehydration in plant biology [11]. In terms of the other metallic system mentioned, sodium sulfite reacts chemically with dissolved oxygen to form sodium sulfate [12]. However, in addition to sodium's adverse effects on plant biology, it would take a large amount of sulfite to neutralize a corresponding amount of oxygen (with a weight ratio of 8:1), the reaction works best at temperatures close to boiling, and possible byproducts include sulfur dioxide, a toxic and odorous gas [8] [12]. 11     Within non-metallic oxygen scavengers, there is greater variety. One oxygen scavenger already in use for the purpose of improving bio-solar cells is catalase. This common enzyme is found in nearly all living organisms exposed to oxygen and catalyzes the decomposition of hydrogen peroxide to water and oxygen as seen below [13]. 2 H2O2 à 2 H2O + O2 Other oxygen scavengers included in this enzymatic category include glucose oxidase and ethanol oxidase. Glucose oxidase transfers two hydrogens from the -CHOH group of glucose to oxygen, forming gluconic acid and hydrogen peroxide [8] [14] [15] [16] [17]. Because of this peroxide product, glucose oxidase is often coupled with catalase, which then creates water and oxygen. These reactions are seen below [14]. Oxygen scavenging systems using this glucose oxidase and catalase combination would need a constant supply of glucose and be sensitive to changes in pH and temperature, among other variables [8]. Because gluconic acid is formed as a byproduct, the solution pH of a closed system can be disturbed considerably [14]. Ethanol oxidases function similarly in that they oxidize ethanol to acetaldehyde, in the process producing hydrogen peroxide that must be neutralized by catalase [8] [18]. Ethanol oxidase/catalase systems would need to be supplied with ethanol to be effective [8]. Ascorbic acid, also known as vitamin C, is a non-enzymatic oxygen scavenger whose method is based on ascorbate oxidation to dehydroascorbic acid [8]. Food packaging systems 12     sometimes include copper to act as a catalyst for the process, as seen in the reactions below [8], where ascorbic acid is abbreviated as AA and dehydroascorbic acid is abbreviated as DHAA, AA + 2 Cu2+ à DHAA + 2 Cu2+ + 2H+ 2 Cu+ + 2 O2 à 2 Cu2+ + 2 O22 O2- + 2 H+ + Cu2+ à O2 + H2O2 + Cu2+ H2O2 + Cu2+ + AA à Cu2+ + DHAA + 2 H2O (or, summarily:) AA + 0.5 O2 à DHAA + H2O but the reaction also naturally takes place in the biological environment [19]. Ascorbic acid provides ascorbate ions that react rapidly with superoxides, singlet oxygens, ozone, and hydrogen peroxide to neutralize these active oxygen species [19]. In addition to this, ascorbate ions help regenerate antioxidant vitamin E and carotenoid pigments used in photosynthesis [19]. Studies in plant physiology show that the monodehydroascorbate free radical oxidized from ascorbate is coupled with ATP formation in the light dependent electron transport from water (recall Figure 1) [20]. In the presence of ascorbate, the net oxygen exchange balance is zero, while the synthesis of ATP increases two to three times due to the extra electron transport to the monodehydroascorbate molecule [20]. Additionally, studies using ascorbic acid as an oxygen scavenger in yoghurt products show no perceivable change in pH while a notable decrease in oxygen and hydrogen peroxide content [21]. Activated carbon, or activated charcoal, is a more mechanical approach to the goal of oxygen scavenging. Activated carbon is a highly porous substance that attracts and holds impurities through adsorption to its high surface area [22]. It has often been used as a cheap, 13     cost effective filtration method for both gaseous and liquid systems [22] [23]. The powdered activated carbons are also easy to use in lab studies due to their non-reactive nature and easy suspension in liquids [24]. Although frequently used to filter out large organic molecules, not much study has gone into investigating activated carbon as an oxygen scavenger. This raises some interesting questions. Activated carbon adsorption of oxygen species would not be selective, and the porosity may capture other necessary organic molecules, hindering thylakoid function. At the same time, the high surface area provided by the carbon particles may not hinder thylakoid function and may even increase DET with the electrode, similar to carbon quantum dot studies discussed earlier [6]. 2.5 RESULTS OF LITERATURE REVIEW Thus, after researching various oxygen scavengers, it was decided that ascorbic acid, based on its multiple recommendations of being beneficial to plant biology and being an effective oxygen scavenger, as well as activated carbon, an additive of unknown but possibly beneficial oxygen scavenging quality, would be tested. These two investigated systems were compared to blank thylakoids with no oxygen scavenger additives and control electrodes imbued with catalase. 14     3. DESIGN PLAN AND PROCEDURE The Design Plan and Procedure section is organized as follows. First come descriptions of the solutions commonly used throughout the project. All component materials were acquired from Sigma Aldrich, and the chemical names of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid and ethylenediaminetetraacetic acid are shortened to their respective acronyms of HEPES and EDTA. Following these come descriptions of the various procedures used throughout the project. For ease of reading, these are listed in the form of numbered steps, with following paragraphs detailing exceptions or additional explanations as needed. The first procedure details the steps included with extracting thylakoids from their source spinach. This is followed by a description of how the chlorophyll content of each batch was measured prior to storage. Chlorophyll concentration per batch is an important quantity because it allows normalization of photocurrent results so that the photocurrents measured from different thylakoid batches, which may have slightly different chlorophyll concentrations (and thus different photosynthetic ability), may be compared. It also allows thylakoid concentration to be estimated. After this, the preparation of thylakoid bio-anodes and laccase bio-cathodes are described, and finally, the procedures behind the two testing methods are listed. Anodes of various compositions were first tested via amperometric testing. This means the photocurrent generated by the electrodes was recorded as a function of time. The purpose of this first mode of testing was to determine which of the tested conditions for each type of oxygen scavenger was most successful at increasing lifetime, as well as to compare the lifetimes of experimental electrodes with those of the blank and control catalase conditions. The second testing method took blank, control, and the longest living experimental thylakoid anodes, connected them with laccase biocathodes to form bio-solar cells, and conducted linear sweep voltammetry with them. 15     This means that the currents produced by the cells were measured across a specified range of voltages to obtain open circuit voltages, short circuit currents, and other measures of solar cell effectiveness to compare the experimental conditions' solar cell performances with those of the blank and control conditions. 3.1 LISTS OF MATERIALS IN COMMONLY USED SOLUTIONS Chloroplast Isolation Buffer (CIB) 5M for 1L, pH = 7.8 1.65 M Sorbitol.........................................................................................................................300 g 0.25 M HEPES.........................................................................................................................59.6 g 50 mM NaCl..............................................................................................................................2.9 g 13 mM EDTA............................................................................................................................3.7 g 53 mM MgCl2...............................................................................................................................5 g Thylakoid Resuspension Buffer for 250mL, pH = 7.8 330 mM Sorbitol....................................................................................................................15.03 g 50 mM HEPES.........................................................................................................................2.89 g 2 mM MgCl2*6H2O...............................................................................................................0.102 g Citrate Buffer for 1L, pH = 5.5* 0.1 M Sodium citrate dihydrate.............................................................................................29.41 g 0.1 M Citric acid....................................................................................................................21.01 g *Separate component solutions mixed at a rough ratio of 2:1 to obtain desired pH 16     3.2 EXTRACTION OF THYLAKOIDS FROM SPINACH 1. Dilute CIB to 1M concentration in advance and allow to chill at 40 oF. 2. Remove the ribs from a bunch of Whole Foods spinach (spinach was obtained consistently from the same store to reduce source inconsistencies) by tearing or cutting. 3. Rinse the torn leaves well with ultrapure water and spin in a salad cleaner to remove excess water. 4. Put the cleaned spinach into the blender with 500 mL of diluted CIB buffer. 5. Blend together at maximum speed for five 1 second pulses. 6. Filter the resulting fluid into two 300 mL beakers through three layers of cheesecloth until both beakers have an equal amount of fluid strained of the largest debris. 7. Pour the equal amounts of fluid into two large centrifuge beakers, then centrifuge at 200g of centrifugal force for 3 minutes at 4 oC to remove any remaining debris. 8. Pour the supernatant into two more large centrifuge beakers and spin at 1000g for 7 minutes at 4 oC. 9. Pour out the supernatant and resuspend the pellets on the bottoms of the centrifuge beakers in 2 mL each of the diluted CIB using a paintbrush. 10. Carefully overlay the suspensions from each beaker into two new, medium centrifuge vials, each containing a mixture of 4 mL Percoll and 6 mL diluted CIB. 11. Centrifuge at 1700g for 6 minutes at 4 oC to obtain whole chloroplasts at the bottom of the vials. 12. Carefully siphon out the supernatant and then resuspend the pellet of chloroplasts in each vial with 20 mL of 2mM MgCl2 to lyse the chloroplasts. 13. Centrifuge at 1700g for 6 minutes at 4 oC. 17     14. Carefully siphon out the supernatant and then resuspend the pellets in each vial in 2 mL each of thylakoid resuspension buffer. 15. Combine both resuspensions in one vial. 3.3 MEASUREMENT OF CHLOROPHYLL CONCENTRATIONS AND STORAGE 1. Set the spectrometer wavelength to a value of 652 nm. 2. Zero the spectrometer at this wavelength with a cuvette filled with 990 µL of 80% acetone solution. 3. Once the spectrometer has been zeroed, tip the vial of thylakoid solution upside down several times to ensure homogenous concentrations of thylakoids throughout the vial, then add 10 µL of thylakoid solution to the cuvette containing the acetone solution. 4. Disperse the thylakoids throughout the cuvette and measure absorbance with the spectrometer. 5. The concentration of chlorophyll in the solution in mg/mL can be found by multiplying the absorbance at 652 nm by 100 (the dilution factor) and dividing that number by 36 (the extinction coefficient). 6. If the overall concentration is greater than 0.2 mg/mL, use thylakoid resuspension buffer to dilute the thylakoid solution the proper amount, then repeat steps 2-5. 7. Repeat step 6 as necessary. 8. Once the thylakoid solution has reached the proper concentration, flip the vial upside down to ensure homogenous thylakoid concentrations and pipet 2 mL into a 2 mL aliquot. 18     9. Repeat step 8 until all thylakoid solution has been pipetted into 2 mL aliquots and then store these aliquots in the freezer at 20 oF for up to a month to be unfrozen as new electrodes as needed. 3.4 PREPARING ANODES For blank electrodes with thylakoid solution alone: 1. Turn on the hot plate to 185 oF to begin melting wax. 2. Turn on the spectrometer and set out a 2 mL aliquot of frozen thylakoid solution to thaw in a room temperature water bath. 3. While the wax is melting, cut out a number of Toray carbon paper electrode bases with heads an area of 1 cm2. Cut enough for at least 2 plain carbon electrodes and at least one extra test electrode for each test condition (if making only one test condition, as in the case of the blanks or the controls, four test electrodes are needed: three to allow triplicate tests and one as a backup). 4. Once the wax has melted, use tweezers to dip the stems of the carbon paper electrodes in the wax until only the heads are uncovered, then place the sticky wax ends on an overturned weigh boat and allow to dry. 5. While the electrodes are drying, turn the thawed aliquot of thylakoid solution over several times to ensure homogenous mixing of thylakoids and measure the concentration of chlorophyll in the aliquot. If the concentration is not within 0.02 mg/ml of the desired 0.2 mg/mL mark, dilute the thylakoid solution with the appropriate amount of thylakoid resuspension buffer. 19     6. Once assured that the thylakoid concentration is within 0.02 mg/mL of 0.2 mg/mL, overturn the aliquot several times to ensure homogenous mixture, then pipet 25 µL of solution onto the head of an electrode and spread evenly across its surface. 7. Repeat step 6 until all the electrode heads have been covered in thylakoid solution but the plain carbon electrodes. 8. Place the weigh boat to which all the electrodes are attached underneath a fan and let dry for 1 hour. 9. Use tweezers to remove the electrodes from the weigh boat and place them face-up in a petri dish. 10. Place the petri dish in a fume hood and put a cap filled with 200 µL of tetramethyl orthosilicate (TMOS) in the center of the petri dish so that all parts of the electrodes remain uncovered. 11. Cover with the petri dish lid and let vapor deposition of TMOS occur for 20 minutes. 12. Remove the electrodes and store them in the dark in the refrigerator to cure overnight at 40 oF. For catalase control electrodes, this procedure is exactly the same except that once the concentration of thylakoid solution has been determined to be acceptable, 0.5 mL of it are pipetted to a separate aliquot and mixed with 15 µL of catalase enzyme in solution form. This mixture is then pipetted onto the electrode heads. Catalase from Aspergillus niger acquired from Sigma Aldrich was used. For the activated carbon electrodes, various suspensions were made of different concentrations of activated carbon in thylakoid resuspension buffer. As no previous work 20     incorporating activated carbon into electrodes had been done, the initial concentrations of these source solutions were 1M, 5M, and 10M, based on sources that used activated carbon for other purposes in a range of concentrations from 1M to 29M [25] [26]. (Note: activated carbon itself has no standard molar weight, so these molar measurements are based on 12.01 g/mol, the molar mass of elemental carbon.) These suspensions were allowed to soak to let buffer displace the air in the activated carbon pores. Once the thylakoid solution concentration had been ascertained, 0.5 mL of thylakoid solution were placed into three new separate aliquots and the soaked activated carbon suspensions were vibrated to resuspend the activated carbon evenly using a vortexer. After this, 15 µL of each concentration was then added to its respective thylakoid aliquot, and the three mixtures were vibrated before each 25 µL were pipetted onto an electrode head to ensure a homogenous suspension of activated carbon particles. Later concentrations of activated carbon source solutions included 0.25M, 0.5M, and 2M. Initially, the production of ascorbic acid electrodes was approached much the same way, with initial investigation showing some studies using as little as 1mM of ascorbic acid in oxygen scavenger tests and other resources showing that levels as high as 10-25 mM of ascorbic acid were naturally present in plant chloroplasts [20] [21]. Thus, similar to the procedure used to make activated carbon electrodes, three initial ascorbic acid solutions in these concentrations were made and later added in small amounts to separated quantities of thylakoid solution before deposition onto the electrode surfaces. However, results using this technique showed no increase from the lifetimes of the blank thylakoid electrodes and raised concerns over whether the acidity of the initial ascorbic acid solutions were harming the thylakoids. Eventually, it was determined that the concentration of ascorbic acid within each electrode was simply too small to have much effect on thylakoid performance. To remedy this, SEM images were taken of thylakoid electrode 21     heads to determine the volume of the dried thylakoid layer, and this volume was used to back calculate the appropriate concentrations of the starting solutions to end up with electrodes actually reflecting the desired testing concentrations of 1mM, 10mM, and 25mM. The respective concentrations of these new source solutions were 0.0343M, 0.343M, and 0.858M. 3.5 PREPARING CATHODES 1. Cut out the desired number of Toray carbon paper electrode bases and coat the electrode tails with wax, but stop before applying anything to the electrode heads. 2. For every three desired electrodes, make an enzyme solution by mixing 1.5 mg laccase from Trametes versicolor from Sigma Aldrich with 75 µL of 0.1M citrate buffer used for testing. 3. Weigh 7.5 mg of anthrocene modified multiwall carbon nanotubes (CNT) into a 2 mL Eppendorf tube. 4. Add the enzyme solution to the CNT vial. 5. Vortex the vial for 1 minute and then sonicate the vial for 15 seconds, using tweezers to securely hold the vial without submerging fingers in the sonication water. 6. Repeat step 5 three more times. This repetitive mixing is needed to align the laccase enzyme with appropriate sites on the anthrocene modified CNT to enable effective electron transfer. 7. Add 25 µL of TBAB (tetrabutylammoninum bromide) Nafion solution and vortex/sonicate one more time to the pattern described in step 5. 8. Use a paintbrush to apply one dip of the resulting slurry to each of the Toray carbon paper electrode heads and clean the paintbrush with multiple rinses of water and ethanol. 22     9. Let the electrode dry for several hours before use or storage in the refrigerator at 40 oF. Once stored, the electrodes will be effective for up to a week. 3.6 AMPEROMETRIC TESTING 1. Place an open faced aluminum box on the counter next to a potentiometer and arrange a ring stand and a 250 W Halogen lamp inside so that the light from the lamp will shine directly into the small 20 mL beaker the ring stand is holding. 2. Fill the 20 mL beaker with citrate buffer. 3. Collect a platinum electrode from the nitric acid tank and rinse with DI water before putting in the beaker and connecting to the counter electrode clip of the potentiometer. 4. Collect a Ag/AgCl reference standard and rinse with DI water before putting in the beaker and connecting to the reference electrode clip of the potentiometer. 5. Connect the working electrode clip to the first of the plain carbon electrodes, making sure that the teeth of the alligator clip have pierced the wax layer on the electrode tail and are in contact with the underlying carbon. 6. Arrange all three electrodes in the citrate buffer beaker so that none are touching and the anode head is facing directly towards the lamp. 7. Once everything is situated, turn on the lamp and cover the box opening with felt so that no outside light can reach the electrode. 8. Turn on the potentiometer and use the test software of CH Instruments CHI660E to run the electrode at a potential of 0.45 V for 600 seconds in the light to eliminate surface reactions on the working electrode. 23     9. After this is complete, run the same conditions again, but turn off the light after 100 seconds, turn on again 100 seconds later, and repeat until 600 seconds has passed. 10. Save and examine results to determine whether the plain carbon electrode is exhibiting any change of photocurrent with the light being on or off. (This should not be the case, as it has no thylakoids or other light-dependent catalysts on its surface to cause a change. If a change is observed, it indicates an amount of error that must be subtracted from the photocurrents seen for the actual test electrodes.) 11. Remove the plain carbon electrode from the working electrode clip. 12. Attach the clip to a test electrode. Run for 600 seconds in the light to eliminate surface reactions. 13. Once surface reactions have been eliminated, set the run time for 2 hours and begin testing with the 100 seconds on, 100 seconds off pattern until a 25% reduction in photocurrent from the second, more stable photocurrent has been observed, then stop and save and remove the tested electrode. 14. Repeat steps 12 and 13 until all test conditions have been tested successfully for three rounds of electrodes. 15. Use time to 25% reduction of initial photocurrent as a measure of lifetime and the difference between the currents in dark and light conditions to determine photocurrent. For the 25mM ascorbic acid electrodes, a light pattern of 200 second on, 200 second off was adopted to account for a slower rate of photocurrent stabilization being observed. Keeping the pattern at the 100 second values would have resulted in unstable photocurrent peaks. 24     3.7 SOLAR CELL TESTING 1. Set up the aluminum box, lamp, ring stand, and buffer beaker as described in steps 1 and 2 of the Amperometric Testing section. 2. Attach a laccase bio-cathode to the working electrode clip of the potentiometer, making sure the alligator clip is making contact with the underlying carbon, and submerge the cathode head in the citrate buffer. Be careful not to lift the cathode above the buffer level before intending to replace it, as the laccase activity will reduce upon reexposure to air. 3. Clip the reference electrode alligator clip to the metal of the counter electrode alligator clip, and then attach this counter electrode clip to the stem of a test electrode. Again, ensure contact with the underlying carbon. Make sure the face of the thylakoid anode is directly facing the lamp. 4. Using felt, block outside light from reaching the electrodes. 5. With the lamp off, use the CH Instruments CHI660E testing software to conduct linear sweep voltammetry on the solar cell. The initial potential should be set to 0.8 V leading down to an end potential of 0.001 V (do not set the end voltage to 0 V, as this is unhealthy for the potentiometer) with a scan rate of 0.005 V/s. This first dark scan will eliminate unpredictable surface reactions occurring on the electrodes. 6. Conduct linear sweep voltammetry with the same parameters on the solar cell again with the lamp off and save the IV curve data from this dark scan. 7. Turn the lamp on, conduct linear sweep voltammetry with the same parameters, and save the data from this light scan for the solar cell. 8. Repeat this procedure so that each type of solar cell (blank, control catalase, longest lifetime ascorbic acid condition, and longest lifetime activated carbon condition) has been 25     tested in triplicate. The voltage where the current is 0 A is the open circuit voltage, and the current where the voltage is 0 V is the short circuit current. 26     4. RESULTS AND DISCUSSION 4.1 AMPEROMETRIC RESULTS Table 1 summarizes the amperometric results collected. "Lifetime" refers to the amount of time it took to reach 25% reduction in initial photocurrent, and "Normalized Photocurrent" refers to the fact that each photocurrent has been normalized by the concentration of chlorophyll in mg/mL found for the thylakoid solutions that made each batch of electrodes. Table 1: Amperometric results of blank, control, activated carbon, and ascorbic acid experimental test electrodes Electrode Type Blank Control (Catalase) 0.25M Activated Carbon 0.5M Activated Carbon 1M Activated Carbon 2M Activated Carbon 5M Activated Carbon 10M Activated Carbon 1mM Ascorbic Acid 10mM Ascorbic Acid 25mM Ascorbic Acid Lifetime and Uncertainty in seconds 983 ± 183 1083 ± 283 1117 ± 383 1117 ± 383 1383 ± 483 1450 ± 250 1117 ± 383 983 ± 583 1450 ± 150 1183 ± 183 1267 ± 433 Normalized Photocurrent and Uncertainty in A/(mg/mL) (4.110 ± 0.770) x 10-7 (2.418 ± 0.745) x 10-8 (3.190 ± 0.640) x 10-7 (2.677 ± 0.563) x 10-7 (2.227 ± 0.437) x 10-7 (1.450 ± 0.140) x 10-7 (1.760 ± 0.500) x 10-7 (2.070 ± 0.980) x 10-7 (3.986 ± 0.393) x 10-7 (4.357 ± 0.745) x 10-7 (8.182 ± 5.243) x 10-7 4.2 AMPEROMETRIC RESULTS DISCUSSION Blank, unmodified thylakoid electrodes were found to last an average of 983 seconds before 25% photocurrent reduction and had a relatively high normalized photocurrent of 4.110x10-7 A per mg/ml of chlorophyll. The pattern of switching the light on and off led to clear, well-defined current peaks as shown in Figure 4. 27     Fig 4: Amperometry of a blank thylakoid anode, where sharp increases in peaks indicate turning off the lamp and sharp decreases in peaks indicate turning on the lamp. The addition of catalase from Aspergillus niger in the control bio-anodes extended the average lifetime of the electrodes to 1083 seconds, or 110% of the lifetime of the blanks. However, the average photocurrent of these electrodes reduced drastically to 2.418x10-8 A/(mg/mL). Results for the varying concentrations of activated carbon tested show more interesting results. Figures 5 and 6 graph the lifetimes and normalized photocurrents of these electrodes with respect to their various activated carbon concentrations. 28     Fig 5: Lifetime of activated carbon electrodes peaks around the 2M concentration Fig 6: Photocurrents of activated carbon electrodes much lower than blank electrode values, especially at high concentrations of activated carbon The lifetimes of these electrodes increase with concentration up to 2M, where lifetime is around 148% that of the blank electrodes, before decreasing with further concentration increases. The uncertainty of the amperometry graphs also increases with concentration. This suggests that activated carbon is to some extent successful as an oxygen scavenger, but at higher and higher 29     concentrations, its advantage as an oxygen scavenger must be balanced with its possible light blocking tendencies. The thylakoid solutions with concentrations of activated carbon higher than 0.5M added were noticeably darker than plain thylakoid solution. At high concentrations, the activated carbon may have literally been blocking the lamplight from reaching a large percentage of thylakoids or been providing too many competing electric pathways with that of the electrode surface, which may account for the increasing amount of noise seen in the amperometric graphs in relation to increasing concentrations. Figure 7 shows the relatively clean data produced by a low concentration 0.25M activated carbon anode with the very noisy data made by a 5M activated carbon anode. This increased noise makes it difficult to obtain a steady current signal from the device. Fig 7: Comparison of (a) clean data for a 0.25M activated carbon bio-anode with (b) the much noisier data gained from a 5M activated carbon bio-anode. 30     When looking at the photocurrents achieved by these electrodes, even the smallest added concentration reduces photocurrent to below that achieved by the blank electrodes, and all the other concentrations show photocurrents significantly lower, with a slight rise at higher concentrations. This suggests that the low selectivity of the activated carbon is adsorbing molecules necessary to the thylakoid function along with oxygen byproducts and affecting their efficiency. The slight increase in photocurrent at 5M and 10M concentrations may indicate that there is a slight amount of increased DET due to surface area at these higher concentrations, but not enough to overcome the adverse effects of the carbon's low selectivity. Despite being lower than that of the blanks, however, all the activated carbon electrodes maintained a much higher photocurrent than those demonstrated by the catalase controls. The 1mM ascorbic acid tests match the activated carbon's highest lifetime increase at 1450 seconds (148% of blank lifetime) but without the higher degree of uncertainty. The photocurrent at this most successful ascorbic acid condition is also much less diminished than its activated carbon counterpart. At 3.986x10-7 A/(mg/mL), this average photocurrent is very close to the original 4.110x10-7 A/(mg/mL) exhibited by the blank bio-anodes. Increasing concentrations of ascorbic acid led to both lowering of lifetimes and increasing of photocurrents. In addition to this, the highest concentration (tested at 25mM ascorbic acid) began to show much slower photostabilization rates of its amperometry peaks, resulting in the need for 200 second lamp on and off periods to capture full photocurrent stabilization. Figure 8 shows a comparison of the sharp, short peaks accomplished at 1mM concentrations with the longer wait times, seen in the much greater peak lengths, required for a 25mM electrode. 31     Fig 8: Comparison of (a) the sharp cut-off times seen every 100 seconds in the low concentration 1mM ascorbic acid electrodes with (b) the long 200 second waits required to see full photocurrent stabilization in a high concentration 25mM electrode All of this evidence suggests that at higher concentrations, the ascorbic acid is escaping the electrode into the citrate testing buffer (resulting in less lifetime stability), where it can act as another oxidizer in addition to the anode, thus artificially increasing photocurrent and making photocurrent stabilization take longer due to the increased amount of oxidizing elements present. The amperometry results thus revealed that the 2M activated carbon bio-anodes and the 1mM ascorbic acid bio-anodes supported the longest lifetimes. As such, these two conditions were carried into solar cell testing along with the blank and catalase control bio-anodes, as discussed in the next sections. 32     4.3 SOLAR CELL RESULTS The following table summarizes the results of the solar cell testing. Within the table "AC" and "AA" stand for "activated carbon" and "ascorbic acid" respectively, "OCV" stands for "open circuit voltage," and "SCC" stands for "short circuit current." The open circuit voltage of a solar cell or battery is maximum potential of the device, and short circuit current is the largest current that can be drawn from the cell [27]. Table 2: Solar cell testing results for the blank, the control, and the experimental conditions with the most longevity Bio-anode and Light Condition Blank, Dark Blank, Light Control, Dark Control, Light 2M AC, Dark 2M AC, Light 1mM AA, Dark 1mM AA, Light OCV (V) SCC (A) 0.727 ± 0.001 0.735 ± 0.005 0.714 ± 0.005 0.714 ± 0.013 0.672 ± 0.042 0.668 ± 0.025 0.731 ± 0.001 0.734 ± 0.002 (5.602 ± 0.416) x 10-7 (7.104 ± 0.555) x 10-7 (6.620 ± 1.071) x 10-7 (6.649 ± 1.050) x 10-7 (7.543 ± 2.229) x 10-7 (7.373 ± 1.663) x 10-7 (6.022 ± 0.735) x 10-7 (6.501 ± 0.652) x 10-7 4.4 SOLAR CELL RESULTS DISCUSSION Representative IV, or current-voltage, curves are shown within this section. Across all the solar cells constructed, it was seen that OCV seldom varied except for in the activated carbon condition, although this was accompanied by a corresponding increase in SCC. The blank solar cells had the largest difference between dark and light SCC's as well as the highest SCC with the maximum OCV seen around the 0.735 V range, as illustrated in Figure 9. 33     Fig 9: IV curves for the blank solar cell devices. These devices showed the highest SCC's of any device as well as the greatest photoelectric effect (largest difference between dark and light photocurrents). Fig 10: Catalase control IV curves, while they had the average SCC closest to initial blank values, had very little difference between their dark and light currents. Catalase control curves had the average SCC that was closest to the initial blank values. However, they did not have much separation between dark and light photocurrents, as can be seen in Figure 10. This indicates that the thylakoids were less effective at transferring their photosynthetic electrons from their photosystems to the carbon paper of their electrodes when in the presence of catalase than without any additional oxygen scavenger present. Activated carbon IV curves showed a surprising trend where device dark currents were actually greater than those of the illuminated devices, as illustrated in Figure 11. This trend stayed constant through multiple tests, so it was determined to be valid. One possible explanation for this trend is that the large surface area of the activated carbon particles helps 34     transfer more electrons to the surface of the electrode during dark conditions. However, once the light is turned on and photoelectrons are generated, this surface area provides too many competing pathways for the electrons, and less actually travel to the electrode surface. Additionally, it can be seen from Figure 11 that the activated carbon curves were very noisy. Even though they had even higher SCC's than the blank devices, their uncertainty of measurement was the highest of any device type, making the activated carbon solar cells the most unreliable of the tested devices. Fig 11: Activated carbon IV curves generally showed dark currents higher than light currents and had the highest SCC's of any device. However, their measurement uncertainty was also the greatest of any device. Similar to the catalase control devices, the ascorbic acid solar cells generated SCC's that were very close to the blank devices' in value. Additionally, as can be seen in Figure 12, the ascorbic acid devices had the largest difference between dark and light currents after those of the blank thylakoid devices, indicating that this condition interfered the least with normal 35     Fig 12: The ascorbic acid solar cells created SCC values very close to the initial blank solar cell currents, as well as having the largest difference between light and dark currents of the experimental conditions and the greatest reliability of any device. type. thylakoid function out of any of the control and experimental oxygen scavenging conditions tested. Finally, the ascorbic acid IV curve values held the least uncertainty of any of the control and experimental values, making the ascorbic acid solar cells the most reliable devices tested. From this solar cell testing as well as the earlier amperometric results, it was concluded that the 1mM ascorbic acid thylakoid bio-anodes created the best combination of lifetime and solar cell functionality of the variables tested. This condition increased lifetime to 148% of the blank thylakoid solution, compared to the 110% increase provided by the catalase control, without the photocurrent reduction or high uncertainty seen in catalase and activated carbon measurements. When used as part of a solar cell, these 1mM ascorbic acid bio-anodes came very close to approaching blank thylakoid solar cell performance, again with the lowest level of uncertainty and SCC's comparable to catalase controls. 36     5. FUTURE WORK Future work in this area of research will include an optimization study of ascorbic acid concentrations like the one done for activated carbon concentrations to determine the best concentration for device performance. It will also involve research into how best to immobilize the concentrations of deposited ascorbic acid onto the electrode surface to avoid the issue of bleeding into the surrounding testing buffer. In addition to further research into ascorbic acid, the most successful of the tested alternatives to catalase, future work will involve investigations into other oxygen scavenging species similar to ascorbic acid in that they already exist inside the chloroplast environments with a number of beneficial functions and side effects. Some examples of these other native oxygen scavenging species include superoxide dismutase, ascorbate peroxidase, β-carotene, αtocopherol, and reduced glutathione [28] [29]. 37     6. BUSINESS, SOCIAL AND ETHICAL CONSIDERATIONS Overall, if bio-solar technology can improve efficiency to 10%, it may compete with traditional solid state solar cells and expand the market for solar energy. Not only would this have an impact on cleaning up the atmosphere of carbonaceous byproducts (or at least slow the increase of carbonaceous byproduct concentration), but it could usher in a new market and set of technologies adapted for disposable power sources. Unlike their more efficient solid state counterparts, the bio-solar cells would be light-weight, inexpensive, able to bend more, and nontoxic, making them perfect for portable, disposable applications. Additionally, if bio-solar cells ever become serious competition for the more expensive and efficient solid state solar cells, their use of abundant materials such as plant matter and silica could reduce the dependency of the technology sphere on rare earth elements that are currently used in attempts to gain the highest efficiencies possible. By reducing dependencies on these elements, which are mostly located in third world countries that control supplies and often mine for the materials in environmentally and socially irresponsible ways, the widespread implementation of bio-solar cells could have an important effect on the economies of rare earth element providers, as well as reduce environmental destruction and human loss of life in the attempt to obtain these materials. New economies and job markets may also subsequently form around the production of the cleaner, cheaper bio-solar cells. 38     7. SUMMARY AND CONCLUSIONS The goal of this project was to discover whether oxygen scavengers other than catalase might increase the lifetimes and performance of thylakoid anode bio-solar cells above the current performance standard. Ascorbic acid was chosen for testing based on ample evidence of its oxygen scavenging ability and beneficial effects in plant biology. Activated carbon was chosen as a material that, having not been studied extensively, had theoretical cases made for either improving or worsening device performance. These oxygen scavengers were tested in comparison to blank thylakoid electrodes and control catalase electrodes first by conducting amperometry on anodes to determine any lifetime benefits and then by testing as whole solar cells with laccase bio-cathodes to determine performance parameters, such as short circuit current and open circuit voltage. From the amperometry testing, it was seen that activated carbon has a beneficial effect on lifetime at some concentrations, but at others either does not have enough concentration to absorb great amounts of oxygen or physically obstructs thylakoids from exposure to light. At any concentration, the activated carbon electrodes showed photocurrents below those created by blank thylakoids, indicating that the poor selectivity of activated carbon to oxygen is allowing it to also adsorb organic molecules needed by the thylakoids to function properly. Additionally, increasing concentrations of activated carbon correlate with increasing noise within electrode amperometry data, suggesting the great surface area of the activated carbon is providing too many competing pathways for photoelectron transferal. The lowest concentration of ascorbic acid tested shows lifetimes comparable to the best lifetimes achieved by the activated carbon, but without the high uncertainty and low photocurrents. However, higher concentrations of ascorbic acid tend to leach out of the electrodes and into the testing buffer. Both of these experimental 39     conditions outperform the blank thylakoid anodes and control catalase thylakoid anodes in terms of lifetime. Solar cell testing revealed that all conditions had comparable open circuit voltages. The catalase solar cells had short circuit currents closest to the ideal blank thylakoid values, but very low differences between dark and light photocurrents indicated that catalase was interfering with thylakoid electron transfer to the electrodes. Activated carbon solar cells showed dark currents that actually had higher short circuit currents than their corresponding light currents. This may possibly be due to the competing pathways that the activated carbon's high surface area would provide for the photoelectrons. Ascorbic acid was again revealed as the most stable alternative to the blank thylakoids, with similar short circuit currents, low uncertainty, and the greatest difference between dark and light currents of any control or experimental condition tested. Thus, low concentrations of ascorbic acid were found to be the most effective oxygen scavenging system tested. The 1mM ascorbic acid condition increased device lifetime to 148% of blank lifetime, compared to catalase's previous standard of only 110%. Additionally, it consistently provided high photocurrents, low uncertainty measurements, and no evidence of interference with thylakoid performance. Activated carbon, despite its advantage in lifetime at some concentrations, can be strongly argued against for applied use due to its negative effect on photocurrent. Both outperformed catalase in terms of lifetime and photocurrent values. 40     8. REFERENCES [1] N. S. Lewis and D. G. Nocera, "Powering the planet: Chemical challenges in solar energy utilization," PNAS, vol. 103, no. 43, pp. 15729-15735, Oct. 2006. [2] M. Rasmussen and S. Minteer, "Photobioelectrochemistry: Solar Energy Conversion and Biofuel Production with Photosynthetic Catalyst," ECS Electrochemistry Letters, vol. 161, no. 10, pp. H647-H655, Jul. 2014. [3] M. Rasmussen et al., "Comparative study of thylakoids from higher plants for solar energy conversion and herbicide detection," Electrochimica Acta, vol. 140, pp. 304-308, Feb. 2014. [4] K. H. Sjoholm et al., "Bio-Solar Cells Incorporating Catalase fro Stabilization of Thylakoid Bioelectrodes during Direct Photoelectrocatalysis," ECS Electrochemistry Letters, vol. 1, no. 5, pp. G7-G9, Aug. 2012. [5] M. Rasmussen and S. Minteer, "Thylakoid direct photbioelectrocatalysis: utilizing stroma thylakoids to improve bio-solar cell performance," Phys. Chem. Chem. Phys., vol. 16, pp. 1732717331, Jul. 2014. [6] M. Rasmussen et al., "Improved Performance of a Thylakoid Bio-Solar Cell by Incorporation of Carbon Quantum Dots," ECS Electrochemistry Letters, vol. 3, no. 2, pp. H1-H3, Dec. 2013. [7] J. Kerry and P. Butler, "Active Packaging of Food" in Smart Packaging Technologies for Fast Moving Consumer Goods, 1st ed., West Sussex, England: John Wiley and Sons Ltd., 2008, ch. 1, sec. 2, pp. 4-6. [8] R. Cruz et al., "Oxygen Scavengers: An Approach on Food Preservation" in Structure and Function of Food Engineering, 1st ed., Intech, 2012, ch. 2, sec.1-2, pp. 21-33. [9] J. Morissey and M. L. Guerinot, "Iron Uptake and Transport in Plants: the good, the bad, and the ionome," Chem. Rev., vol 109, no. 10, pp. 4553-4567, Oct. 2009. [10] Lenntech, (2016). Fenton's reaction [Online]. Accessed Feb. 12, 2016. Available: http://www.lenntech.com/fenton-reaction.htm [11] B. L. Grant, (2015). Sodium Tolerance of Plants- What Are the Effects of Sodium in Plants? [Online]. Available: http://www.gardeningknowhow.com/garden-how-to/soil-fertilizers/sodiumin-plants.htm [12] Guardian CSC, (2016). Oxygen Scavengers [Online]. Accessed Feb. 12, 2016. Available: www.guardiancsc.com/pdf/5_02_scavengers.pdf 41     [13] P. Chelikani et al., "Diversity in structures and properties among catalases," Cellular Molecular Life Sciences, vol. 61, no. 2, pp. 192-208, Jan. 2004. [14] S. Kim et al., "The Effect of Oxygen Scavenging System on the pH of Buffered Sample Solutions: in the Context of Single-Molecule Fluorescence Measurements," Bull. Korean Chem. Soc., vol. 33, no. 3, pp. 958-962, Jan. 2012. [15] (2012). Enzyme-based oxygen scavengers based on glucose oxidase/catalase [Online]. Accessed Feb. 12, 2016. Available: http://www.foodtech-portal.eu/index.php?title=Enzymebased_oxygen_scavengers_based_on_glucose_oxidase/catalase [16] V. Leskovac et al., "Glucose oxidase from Aspergillus niger: the mechanism of action with molecular oxygen, quinones, and one-electron acceptors," Int. Journal of Biochem. and Cell Bio., vol. 37, pp. 731-750, 2005. [17] S. Bankar et al., "Glucose oxidase - An Overview," Biotech. Advances, vol. 27, pp. 489501, Apr. 2009. [18] P. Goswami et al., "An Overview on Alcohol Oxidases and their Potential Applications," Appl. Microbiol. Biotechnol., pp. 1-17, Mar. 2013. [19] G. Potters et al., "Ascorbate and glutathione: guardians of the cell cycle, partners in crime?," Plant Physiol. Biochem., vol. 40, pp. 537-548, Mar. 2002. [20] G. Forti and G. Elli, "The Function of Ascorbic Acid in Photosynthetic Phosphorylation," Plant Physiol., vol. 109, pp. 1207-1211, Aug 1995. [21] R. Dave and N. P. Shah, "Effectiveness of Ascorbic Acid as an Oxygen Scavenger in Improving Viability of Probiotic Bacteria in Yoghurts Made with Commercial Starter Cultures," Int. Dairy Journal, vol. 7, pp. 435-443, Apr. 1997. [22] General Carbon Corportation, (2016). Cleaning the World with Activated Carbon [Online]. Accessed Feb. 12, 2016. Available: http://generalcarbon.com/facts-about-activatedcarbon/activated-carbon-faq/ [23] Chemistry Explained, (2016). Carbon [Online]. Accessed Feb. 12, 2016. Available: http://www.chemistryexplained.com/elements/A-C/Carbon.html [24] (2014). Which type of activated carbon do you recommend for gas adsorption applications - granular or powdered? [Online]. Accessed Feb. 12, 2016. Available: https://www.researchgate.net/post/Which_type_of_activated_carbon_do_you_recommend_for_g as_adsorption_applications-granular_or_powdered 42     [25] G. Drazer et al., "Concentration dependence of diffusion-adsorption rate in activated carbon," Chem. Eng. Sci., vol. 54, pp. 4285-4291, Jan 1999. [26] J. Racyte et al., "Effect of Granular Activated Carbon Concentration on the Content of Organic Matter and Salt, Influencing E. coli Activity and Survival in Fluidized Bed Disinfection Reactor," Biotechnol. and Bioeng., vol. 111, no. 10, pp. 2009-2018, Oct. 2014. [27] C. Brett and A. Brett, "Electrochemical Cells: Thermodynamic Properties and Electrode Potentials" in Electrochemistry: Principles, Methods, and Applications, 1st ed., Oxford, England: Oxford University Press, Inc., 1993, ch. 2, pp. 13-38. [28] K. Asada, "Production and Scavenging of Reactive Oxygen Species in Chloroplasts and Their Functions," Plant Physiol., vol. 141, no. 2, pp. 391-396, Jun. 2006. [29] K. Das and A. Roychoudhury, "Reactive oxygen species (ROS) and response of antioxidants as ROS-scavengers during environmental stress in plants," Front. Environ. Sci., vol. 2, no. 53, pp. 1-13, Dec. 2, 2014. 43