Study of Arsenic Biosorption in Ascophyllum Nodosum
Full text
Integrated Master on Environmental Engineering 2014/2015 Study of arsenic biosorption in Ascophyllum nodosum Ana Carolina Duarte Ramos Ferreira Master thesis developed in order to get the title of ENVIRONMENTAL ENGINEERING MASTER President of the committee: Manuel Fernando Ribeiro Pereira (Course Director, and president of the Scientific and Course accompaniment commissions for the Environmental Engineering Department) Academic supervisor: Sílvia Cristina Rodrigues dos Santos (Post-doctoral researcher of the Chemical Engineering Department) Academic co-supervisor: Cidália Maria de Sousa Botelho (Auxiliary professor and Council member of the Chemical Engineering Department Scientific commission and Course accompaniment commission member for the Environmental Engineering Department) January 2015
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 2 Acknowledgments I would like to express my most sincere gratitude to professors Sílvia Santos and Cidália Botelho, for the motivation and guidance they provided me, as well as valuable ideas, during the time it took to complete this work. I would also like to thank my laboratory colleagues and friends, which facilitated my setting to a new work environment and provided me answers every time they could. In particular, I’d like to acknowledge my friend and master thesis companionship, Dulce, for the friendship and help along the project. In addition, a “thank you” for my university, FEUP, and all the professors that accompanied me along these five years, for providing me the conditions to enjoy a fantastic education system, and the motivation to learn everyday a little more, in a particular way to LSRE which provided me with the means and equipment to write this thesis. My gratefulness goes, also, to my friends that supported me in various ways, and helped me survive the stress of weeks and months of laboratory work, above all to those who have a special role in my life, and had been with me for most of my life. To my brother, for the days when I was not so easy to deal with, and for the reassuring words. To my boyfriend, Ricardo, for all the hours spent helping me and cheering me up, for the great support you gave me all those years, and for that unconditional love and friendship we share. But most of all, I dedicate this thesis to my parents who have always set me towards the higher targets and gave me their full confidence and care. For the belief they have in me and the strength they offer me for always pursuing my dreams. For accepting all my flaws and still love me with no restrictions, and for the patience and education. For them, I have nothing but a profound gratitude and an even deepest love.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 3 Abstract The present project was intended to study the viability of Ascophyllum nodosum in the removal of As(V) from contaminated waters through an adsorption process. This algae was used in its natural form, although a calcium treatment was also tested in order to positively charge the algae surface and thus improve the adsorption capacity of the algae. This treatment was done with and without a previous protonation of the seaweed. This particular seaweed was chosen due to its large availability on the Portuguese shore, and based on several research studies regarding adsorption of cationic metals on this species, that have proven to be a low cost option to the commercial adsorbents. The calcium treatment proposed has shown that the calcium content increased to 35-36 mg/g in both protonated and non-protonated seaweed. Calcium adsorption isotherms were previously determined. Langmuir model presented a higher adsorption to the virgin algae with a Qmax of 419 ± 384 mg/g. This value was not however obtained in the final treated seaweeds due to the calcium loss during the washing steps after the treatment. Surface charge of virgin and calcium treated seaweed was proven consistent with the goal of the treatment, with the calcium treated seaweed presenting a positive surface charge curve within most of the pH range studied. Natural and calcium-treated algae were studied as biosorbents for As(V). The biosorption dependency to pH was evaluated and better values were obtained for pH 5-6. For better understanding of the process, isotherm assays for arsenic adsorption were made and adjusted to Freundlich and Langmuir’s models. The model that best adjusted to data was Langmuir’s. For the arsenic adsorption assay the virgin algae was also the one with higher results, presenting a Qmax of 200 ± 86 µg/g. It was also made a kinetics study regarding the sorption of As(V) to calcium treated algae. Equilibrium was reached after the 2 hours contact time. The most of the times the virgin algae presented higher or similar values of adsorption which may mean that the treatment is not necessary.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 4 Table of Contents 1. Introduction ........................................................................................................................................................... 6 1.1. The Problem ................................................................................................................................................ 6 1.1.1. The Goal ............................................................................................................................................. 6 1.2. Arsenic ......................................................................................................................................................... 7 1.3. Ascophyllum nodosum ............................................................................................................................... 10 1.4. Existent Methods of Treatment .................................................................................................................. 11 1.4.1. Coagulation/Flocculation or Chemical Precipitation .......................................................................... 11 1.4.2. Ion Exchange .................................................................................................................................... 11 1.4.3. Microfiltration, Nanofiltration, Ultrafiltration and Reverse Osmosis.................................................... 12 1.4.4. Adsorption/Biosorption ...................................................................................................................... 12 2. Materials and Methods ....................................................................................................................................... 15 2.1. Chemicals .................................................................................................................................................. 15 2.2. Equipment .................................................................................................................................................. 15 2.3. Analytical Methods ..................................................................................................................................... 16 2.4. Experimental Procedure............................................................................................................................. 17 2.4.1. Algae ................................................................................................................................................. 17 2.4.2. Algae Treatment ................................................................................................................................ 17 2.4.3. Calcium Adsorption Isotherm ............................................................................................................ 19 2.4.4. Chemical Characterization ................................................................................................................ 19 2.4.4.1. Calcium Quantification ...................................................................................................................... 19 2.4.4.2. Potentiometric Titrations .................................................................................................................... 20 2.4.5. Effect of pH in As adsorption process ............................................................................................... 21 2.4.6. Arsenic Adsorption Isotherm ............................................................................................................. 21 2.4.7. Arsenic Adsorption Kinetics ............................................................................................................... 22 3. Results and Discussion ...................................................................................................................................... 23 3.1. Calcium Adsorption Isotherm ..................................................................................................................... 23 3.2. Chemical Characterization ......................................................................................................................... 26 3.2.1. Calcium Quantification ...................................................................................................................... 26 3.2.2. Potentiometric Titrations .................................................................................................................... 27 3.3. Arsenic Adsorption Tests ........................................................................................................................... 29 3.4. Arsenic Adsorption Isotherm ...................................................................................................................... 32 3.5. Arsenic Adsorption Kinetics ....................................................................................................................... 35 4. Conclusions ........................................................................................................................................................ 37 5. Suggestions for Future Work .............................................................................................................................. 38 6. References ......................................................................................................................................................... 39 7. Attachments ........................................................................................................................................................ 45
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 5 Figure Index Figure 1 - Arsenic distribution in Portugal on the soil (left) and on water (right) (from [18]) ........................... 8 Figure 2 - Arsenic speciation with pH ............................................................................................................ 8 Figure 3 - Ascophyllum nodosum .................................................................................................................10 Figure 4 - Ascophyllum nodosum granulometry used in the project (dry algae) ...........................................17 Figure 5 – Experimental data obtained from the calcium adsorption isotherm assay. Experimental conditions: V = 50 mL; m = 0.5 g; T = 25 C; t = 24 h; v = 180 rpm; pH 6 ± 0.5; CCa = [0.002 M; 1 M]. .........23 Figure 6 – Ajdjusts for Av in the calcium adsorption isotherm assay. ...........................................................25 Figure 7 – Adjusts for Ap in the calcium adsorption isotherm assay. ...........................................................25 Figure 8 – Calcium quantification in seaweed, for the different treatments used on the project. ..................26 Figure 9 - Experimental data obtained for the surface charge of both algae. Experimental conditions: T = 25 °C; electrolyte: 0.1 M NaOH solution; ..........................................................................................................28 Figure 10 - Experimental data obtained for the pH influence on arsenic adsorption assay. Experimental conditions: V = 50 mL; m = 0.35 g; t = 6 h; v = 180 rpm; CAs = 25 mg/L ......................................................29 Figure 11 - Experimental data obtained from the arsenic adsorption isotherm assay. Experimental conditions: V = 50 mL; m = 2 g; T = 25 C; t = 8 h; v = 180 rpm; pH 6 ± 0.5; ................................................32 Figure 12 - Adjusts for Av in the arsenic adsorption isotherm assay. ...........................................................33 Figure 13 - Adjusts for ACa in the arsenic adsorption isotherm assay .........................................................34 Figure 14 - Experimental data obtained from the arsenic adsorption kinetics assay. Experimental conditions: V = 500 mL; m = 20 mg; t = 8 h; v = 200 rpm; pH 6 ± 0.5; CAs = 25 mg/L. ................................35 Table Index Table 1 – Maximum adsorption capacities (Qmax) reported for arsenic removal in aqueous solutions (adapted from [42]) ......................................................................................................................................13 Table 2 - Nomenclature and colour pattern used during the project to identify the algae .............................18 Table 3 - Precipitation prediction of calcium arsenate through its Ksp ..........................................................31 Table 4 - Precipitation prediction of calcium hydroxide through its Ksp .........................................................31 Table 5 - Langmuir adjust parameters for Av (value ±interval for 95% confidence). ....................................34
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 6 1. Introduction 1.1. The Problem Heavy metals and similar elements, as metalloids, are a big concern, considering its high toxicity. These elements should be treated when present in waste water and groundwater, because they can be a threat to public health when consumed above certain concentrations. In this case of study, the goal is to pursue new low cost treatments, particularly biosorption by seaweeds, for the metalloid arsenic (As). This type of treatment has already been extensively studied for different metals and other elements that behave similarly, and has shown good results [1,2]. But, as most metals studied in aqueous solutions are cations, the process for removing As, a highly toxic substance, is quite unknown, since it takes the form of oxyanion in aqueous solution, and the process behind their biosorption is not well studied yet. 1.1.1. The Goal There are already some effective treatments for the removal of the studied metalloid, which are presently used in most Water Treatment Plants, such as ion exchange, micro and ultrafiltration, chemical precipitation and even adsorption, usually with inorganic, commercial materials [3]. In this project, it is aimed to find a new, more economical, way of treatment, using a low cost organic material as adsorbent (in this case, the seaweed that can be commonly found lying on the Portuguese shore most of the year), and access the cost-benefit proportion, and the pre-treatments that may be necessary, or not, to optimize the removal capacity for this element.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 7 1.2. Arsenic Arsenic occurs in the Earth’s crust in more than 200 minerals. Its concentration in soil is between 0.2 and 40 mg/kg when the soil is not contaminated, or between 100 and 2500 mg/kg in the contaminated ones. In natural waters, the As concentration range goes from 0.5 μg/L to more than 5000 μg/L, although the typical concentrations tend to be below 10 μg/L. Arsenic can be available for human exposure from either natural or anthropogenic sources. The presence of this metalloid in water can be due to a natural presence in the local bedrock or from arsenic-rich geothermal fluids that have contact with surface waters. Looking at the anthropogenic sources, the activities that seem to be the most responsible for causing arsenic pollution, are mining, especially areas where sulphide minerals tend to oxidize, or mining related activities, since gold and arsenic-bearing minerals tend to coexist, as well as the use of pesticides [4-6]. Arsenic has been found in higher concentrations in groundwater areas from Argentina, Chile, Mexico, China and Hungary, and recently in India (West Bengal), Bangladesh and Vietnam. In some of these areas, people are exposed to this contamination trough drinking water. [6] In Europe there have been several studies related to arsenic pollution in rivers and water masses. Comparing river, lake and estuarine waters from Norway, France, Italy, Bavaria, Belgium, Yugoslavia and some polluted European rivers the average values varied from 0.02 µg/L to 9.2 µg/L, [8-10] and the Atlantic ocean presented values in the range of 1.1 – 1.8 µg/L [11]. Some naturally occurring As problems in European groundwaters have also been included, such as a range of 2 – 176 µg/L in Hungary and Romania, in the Danube basin and Greece with concentrations as high as 10000 µg/L in underground waters near gold mining activities [12-14]. In France a profound study of the arsenic in Garonne, Dordogne and Isle Rivers, was made where the dissolved As higher values where referenced as 6000 ng/L, which doesn’t seem to be as much as in other referenced places of Europe [7]. In southwest Finland, dissolved arsenic concentrations from 17 µg/L to 980 µg/L have also been found in natural wells [15,16]. In the case of Portugal, much arsenic can derive from the dissolution of sulphur-rich minerals (from the volcanic and geothermal activities), and, as referenced before, from mining. As an example, we can look at Castromil region. This location, in the north of Portugal, is the crib of old, abandoned gold mines since 1940. Since there no environmental regulations have been established back then, this region is now a residential area suffering from high concentrations of arsenic, as well as other toxic elements, such as Pb, in the agricultural soil [17].
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 8 In figure 1, is possible to perceive the arsenic distribution in Portugal, in soils and natural water. Arsenic is more commonly found on two species: trivalent arsenic – As(III), found in reductor environments in the form of arsenite and pentavalent arsenic – As(V), found in oxidizing environments and in the form of arsenate. In the aqueous solution the form in which arsenic is found depends on the amount of oxygen present (redox potential) and pH [19,20]. The arsenic speciation as a function of pH is shown in figure 2 [21]. Figure 2 - Arsenic speciation with pH Figure 1 - Arsenic distribution in Portugal on the soil (left) and on water (right) (from [18])
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 9 On a toxicity level, As(III) is usually more toxic and mobile than As(V), showing more acute health symptoms after exposure, which is usually by ingestion. The difference for acute toxicity tested in mice can go as high as 160 to 165 times [20,22,23]. This toxicity difference between the two arsenic forms can be biologically explained since trivalent arsenic interferes with enzymes responsible for the generation of cellular energy, inhibiting the ATP production. As a result, energy production is reduced damaging the cells slowly leading to cellular death. Pentavalent arsenic, in the inorganic form of arsenate, when inhaled or ingested tends to compete with phosphate which leads to a small interference in the ATP production but not as big as As(III) and can at some point interfere with the DNA chain, but doesn’t necessarily results in cellular death, and as long as there’s phosphate in the organism, arsenate tends to be more easily expelled [24]. According to the review article about sources, behaviour and distribution of arsenic, by Smedley and Kinninburgh [12], it would seem that there’s two triggers that can precede the release of large amounts of As in aquifers. The first would be high pH conditions (pH>8.5) in semi-arid and arid environments, which would lead to the desorption of absorbed arsenic, especially As (V). The second trigger would be the exposure to extremely reducing conditions allied to a neutral pH value, originating the desorption of As (III) from mineral oxides. Human health problems caused by arsenic depend, of course, on the level of exposure to the contaminant. Therefore, an acute exposure to inorganic arsenic by inhalation can cause gastrointestinal complications, such as diarrhea, nausea and abdominal pain, or even central and peripheral nervous system disorders. Some studies show that the inhalation of this compound can be strongly associated with lung cancer incidence, and its ingestion can increase the risk of skin, bladder, liver and lung cancer [25]. Oral exposure in high doses can cause death. If the levels of ingestion are lower than 600 µg/kg/d, it can cause nausea and vomiting and affect the central nervous system (CNS), cardiovascular system, liver, kidneys and blood (anaemia). If the exposure to inorganic arsenic is chronic, it can lead to skin and mucous membranes irritation in lower quantities or, if there are elevated levels of exposure, it can result in gastrointestinal affects, anaemia, skin and vascular lesions, and even neurocognitive and behavioural changes in school age children, among other effects. It has also been proved that, in pregnant women, when ingested, the chemical can cross the placenta, exposing the fetus to it.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 16 2.3. Analytical Methods The main analytical method used in the analysis of the samples was Atomic Absorption Spectrometry (AAS). The equipment atomises the solution, and uses a specific wavelength of light for each element, corresponding to the one known to be better absorbed by the element tested. The result is, therefore, based on the amount of radiation absorbed during the process, and the amount of atoms of the element analysed is proportional to it. In the end, the absorbance is compared to a calibration curve, made earlier in the analysis, to enable the calculation of the element concentration in the unknown sample [60]. The equipments used (Atomic Absorption Spectophotometry and Graphite Furnace Atomic Absorption) differ on the process of atomization. The first uses a flame aspiration process in which, depending on the element to analyse, are used mixtures of gases such as ethyne/air (2200-2400°C flame temperature) and ethyne/dinitrogen oxide (2600-2800°C flame temperature). In this flame process, the solution passes through a nebulizer that “breaks” the sample and allows only the small drops to go through the burner and, therefore, be read. The second process is based on an electrothermal atomization, which occurs in a hollow graphite tube. This tube is electrically heated by a passing current, with a very fast heating rate (2000-2500°C in 5-10 seconds), which vaporizes the elements in the sample, atomizing almost 100% of it. In the end the tube is heated until 2700°C to clean it for the next sample. Along the heating cycle, the tube is flushed with argon gas to prevent it from burning away. This technique is far more sensitive than flame AAS [60]. To the calcium analysis a ethyne/air flame was used. The analysis was performed with a multi-element lamp, using wavelengths of 422.7 and 239.9 nm, a slit width of 0.5 nm and a 5 mA lamp current. On the matrix of the samples depending on the calibration curve chosen 0.6 – 10 mg/L or 10 – 600 mg/L a lanthanum solution (5 g/L) or potassium chloride (4 g/L) was used, respectively to reduce the interference. Arsenic was determined using an ethyne/dinitrogen monoxide flame and a calibration in the range of 3 – 50 mg/L. The analysis was performed using a wavelength of 193.7 nm, a slit width of 1.0 nm and a 5.0 mA lamp current. The calibration lines used in the processes are presented as an attachment to the present document.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 17 For low arsenic concentrations which were not covered by the calibration line, successive dilutions were made, instead of one high dilution, to reduce the dilution error, in order to read those samples in the Graphite Furnace with a wavelength of 217.5 nm, a slit of 0.2 nm, and a 10 mA lamp current along with background correction. 2.4. Experimental Procedure 2.4.1. Algae Ascophyllum nodosum was collected in the northern shores of Portugal, near Viana do Castelo, where this brown seaweed washed ashore in great quantity. This sundried alga was washed later with both tap and distilled water in order to remove salt and sand. The seaweed was washed, several times, until it was detached of most part of the exceeding organic material and its conductivity was near that of the distilled water. Posteriorly it was dried in an oven at 60ºC, shredded in a mill to obtain a lower granulometry (figure 4), and stored [61]. 2.4.2. Algae Treatment The chosen seaweed was used in both treated and virgin forms, the last was referenced as Av. The treatment of the seaweed was primary based on the procedure referred by Costa et al. [61] and it aimed to create calcium bindings at the surface of the seaweed, so that it becomes a positive charged surface enabling the connections with the anions we want to remove from the solutions. It was based on a protonation, which targets the saturation of the surface of the organic material with the H+ ion, facilitating the trade of the 1cm Figure 4 - Ascophyllum nodosum granulometry used in the project (dry algae)
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 18 ion H+ with Ca2+, so that the second one can fix on the algae surface, by a treatment with a calcium solution. The seaweed protonation was made using 40 g of Av, and immersing it on 4 L of HNO3 0.2 M solution which was set under mechanical agitation (200 rpm) for 3 hours. This procedure was repeated after settling of the seaweed in the bottom of the recipient, to ensure the minimal loss of the biomass when switching from the old to the new solution. After this repetition, the seaweed was washed with distilled water until its pH was near to the range of 4 – 5. For that treated Ascophyllum nodosum was chosen the acronym Ap. The first calcium treatment tested (also based on [61]), was made after draining the protonated algae. This material was placed into 4 L of a CaCl2 0.2 M solution, under the same type of stirring for 24 hours. The next step consisted in washing the seaweed as many times as needed to guarantee that the conductivity of the washing water was close to that of distilled water, therefore assuring that the only remaining calcium was the one attached to the biomass surface. To this treated seaweed was given the acronym ApCa. Later in the project, the protonation phase of the treatment was dropped and the adsorbent suffered the same calcium treatment with a difference on the concentration of the calcium solution that was switched to CaCl2 0.75 M. This change was made, based on the results of the calcium adsorption isotherm, presented on the 3.1 section of the present document. This alga was named ACa. The following table summarizes the acronyms given to each treatment and the colour that will represent this seaweed forms in the graphics, along the document. Table 2 - Nomenclature and colour pattern used during the project to identify the algae Acronym Treatment Colour Av Non-existent Orange Ap Protonation Purple ApCa Protonation + CaCl2 0.2M solution Green ACa CaCl2 0.75M solution Blue After each treatment, the seaweed was dried in the oven, at 60 °C for approximately 24 hours, and weighted to verify the loss of biomass in these processes.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 19 2.4.3. Calcium Adsorption Isotherm Next to the preliminary phase of the project, a calcium adsorption isotherm was determined to discover the maximum concentration of calcium that could be attached to the surface of the seaweed and the initial Ca concentration that should be used. This procedure was done by exposing 0.5 g of biomass to different calcium solution concentrations, under the same conditions: V = 50.0 mL (in each flask); m = 0.5 g (accurately weighted, in each flask); T = 25 °C; t = 24 h; stirring rate = 180 rpm; pH = 6 ± 0.5. This procedure was made in duplicate for both the protonated and virgin seaweed, and with crescent concentrations of calcium solutions: 0.002 M; 0.005 M; 0.05 M; 0.2 M; 0.5 M; 0.7 M; 0.75 M; and 1 M. After the assay was finished, the samples were filtered and solutions analysed by flame AAS along with the initial solutions to calculate the amount of Ca that the biomass retained from each solution, and which was the equilibrium concentration reached. 2.4.4. Chemical Characterization 2.4.4.1. Calcium Quantification A biomass digestion was made for alga samples Av, ApCa and ACa, in triplicate, and in duplicate for the control digestion – blank (without seaweed). The digestion solution used was aqua regia with the following proportions: 1.0 g of biomass (accurately weighted; excluding the control tubes); 5.0 mL of purified water; 12.0 mL of HCl (37%, analytical gradeSigma-Aldrich); 4.0 mL of HNO3 (68-70%, analytical gradeScharlau);
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 20 The digestion was made in a program of 2 h at 150 °C, than samples were filtered with a cellulose acetate membrane, 45 µm pore size, for 100 mL (Av and ApCa assay) or 50 mL (ACa assay) volumetric flasks and completed with purified water. The calcium concentrations in these solutions were later obtained by flame atomic absorption spectrophotometer (flame AAS). This step was made, in order to realize how much calcium was retained by each seaweed, with the calcium treatment, and if the protonation before the calcium treatment was needed to make a difference in the adsorbed calcium. 2.4.4.2. Potentiometric Titrations A titration is a process in which the concentration/quantity of the analyte (constituent of the sample) is determined by the addiction of known quantities of other substance which is normally known to react in a standard way with the analyte [62]. This particular type of titrations is based on an electric potential difference measurement, which is made using an electrode. This means that, in this method, opposing most of the potentiometry methods, the aim is to find the change in the electrode potential, rather than an accurate value for the electrode potential. Therefore, the progress of the reaction is followed through the differential potential change [62,63]. The goal of this method was to find the pH of zero surface charge of the adsorbent (pHzc), which gives information about the possible attractive-repulsive reaction between sorbent and adsorbate, and enables to ensure that electrostatic forces are important on the metal sorption mechanism. One of the main goals is to study and compare the surface chemistry of the treated and non-treated algae. In each titration 0.250 g of Av or ACa was put in contact with 50.0 mL of electrolyte solution (0.1 M NaCl), under stirring, in nitrogen atmosphere (to avoid interference of CO2) and for 1 hour. A volume (accurately measured, and recorded, about 2 – 5 mL) of 0.1 M HCl solution, previously standardized, was then used to adjust the starting pH to 2.5. When the pH stabilized, the titration started, using NaOH 0.109 M solution as titrant. Additions were automatically made, in increments of 0.02 mL, when the drift in potential measure was lower than 0.5 mV/min or after 20 min. A blank titration (with the electrolyte solution) was also performed.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 21 2.4.5. Effect of pH in As adsorption process A mass/volume ratio of 7 g/L was used, and the assay was done in triplicate for both Av and ApCa samples. This assay was made, controlling the pH of the solutions to the desired value (±0.5), and using different pHs to study the influence of this factor on the arsenic biosorption, maintaining constant the other parameters: V = 50.0 mL (in each flask); m = 0.35 g (in each flask); T = 25°C; t = 6 h; stirring rate = 180 rpm; Initial As concentration = 25 mg/L. The pH’s studied where 3, 4, 5 and 6, since usually arsenic contaminations tend to occur in lower pH environments and therefore it is not cost-effective to use much higher pH’s. After this assay the samples were filtered with a cellulose acetate membrane, 45 µm pore size, and concentrations read on the flame AAS. In the end of ACa, assay, calcium concentration in the liquid phase was also read on flame AAS. The aim was to evaluate the amount of Ca drawn back to the solution and if its concentration is enough to Ca3(AsO4)2 precipitation. 2.4.6. Arsenic Adsorption Isotherm The isotherm study was made in order to acquire better knowledge of the equilibrium concentration between As in the liquid and solid phases, as well as discovering the maximum arsenic concentration that could be removed by adsorption to the surface of the seaweed. This procedure was done by exposing a constant adsorbent dosage to different concentration arsenic solutions, under the same conditions: V = 50 mL (in each flask); m = 2 g (in each flask); T = 25°C; t = 8 h; rpm = 180; pH = 6 ± 0.5.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 22 This procedure was made in duplicate for Av and ACa, and with crescent concentrations of arsenic solutions: 2 mg/L; 5 mg/L; 10 mg/L; 15 mg/L; 25 mg/L and 50 mg/L. After the assay was finished, the samples were filtered with a cellulose acetate membrane, 45 µm pore size, and analysed for As on the flame AAS or AAS-graphite furnace along with the initial solutions. The amount of As that the biomass retained in equilibrium, per unit mass of adsorbent (qe) was calculated by equation (1), where Ci and Ce represent the initial and equilibrium concentrations, respectively, V stands for the volume and m represents the algae mass. 𝑞𝑒=(𝐶𝑖−𝐶𝑒)×𝑉 𝑚 (1) 2.4.7. Arsenic Adsorption Kinetics A kinetic assay was made, in order to discover the behaviour of the arsenic adsorption along time and the time needed to reach the adsorption equilibrium. This procedure was duplicated for both Av and ACa, and it was performed using a seaweed dosage of of 40 g/L. In this process we weighted 20 g of algae for each assay, along with 500 mL of a 25 mg/L arsenic solution, and it was put under agitation (200 rpm) for 8 hours at pH 6 ± 0.5. During the time of the assay, several samples of the mixture were collected, at different time stamps (0 min; 15 min; 30 min; 1 h; 2 h; 3 h 30 min; 5 h; 8 h), and filtered with a cellulose acetate membrane, 45 µm pore size, to be analysed by flame AAS along with the initial solution (t=0min), to determine the arsenic removal in each time stamp.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 23 3. Results and Discussion 3.1. Calcium Adsorption Isotherm The graphic below presents the experimental results for the calcium adsorption isotherm. The variable qe represents the amount of calcium adsorbed per unit mass of adsorbent which in this case is per gram of algae. The Ce variable represents the calcium concentration in the aqueous solution at the equilibrium. This isotherm study was conducted in order to determine the initial concentration needed to origin a larger calcium quantity adsorbed in the algae after the calcium treatment. Figure 5 – Experimental data obtained from the calcium adsorption isotherm assay. Experimental conditions: V = 50 mL; m = 0.5 g; T = 25 C; t = 24 h; v = 180 rpm; pH 6 ± 0.5; CCa = [0.002 M; 1 M]. Observing figure 5 it is possible to see that the concentrations of Ca in the solid phase tend to a constant value, around 250 mg/g, corresponding to calcium concentrations in the initial solution higher than 0.7 M. In this chart we can also conclude that, based on the values and corresponding deviations, the calcium adsorbed quantity is almost the same in both protonated and virgin seaweed, so it is more profitable to use the virgin one, since the protonation requires the use of more reagents. 0 100 200 300 400 010 20 30 40 50 qe(mg/g) Ce×103(mg/L) Av Ap
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 24 Based on this analysis the initial calcium concentration chosen for the algae treatment was 0.75 M. To describe the adsorption equilibria it is usually better to represent it by explicit mathematical equations. The most used equilibrium models are the Langmuir Model and the Freundlich Model, represented by the equations 2 and 3, respectively. [64] 𝑞𝑒=𝑄𝑚𝑎𝑥 × 𝐾𝐿× 𝐶𝑒 1 + 𝐾𝐿× 𝐶𝑒 (2) 𝑞𝑒= 𝐾𝐹× 𝐶𝑒 1 𝑛 (3) Looking at the equations individually, we can identify KL and KF as the Langmuir and Freundlich isotherm constants, respectively, while Qmax is the maximum monolayer coverage capacity and n characterizes the adsorption intensity. The Langmuir Model describes quantitatively the formation of the monolayer on the adsorbent as well as the point after which no more adsorption occurs. This isotherm is only valid for monolayer adsorption to a surface containing finite number of homogeneously distributed binding sites, and it assumes that there are uniform energies of adsorption in the process and no migration of the adsorbate on the surface of the adsorbent. The difference between the two models is that Freundlich Model characterizes de adsorption on a heterogeneous surface, and therefore, different affinity for adsorption [65,66]. The fitting of models to experimental points was made by nonlinear regression (to Eq. 2 and 3), by minimizing the sum of squared residuals, using CurveExpert professional software. However as the obtained deviations were too high, the adjusts produced statistically meaningless results. As the goal for this assay was to determine the higher calcium adsorption, using the lowest concentration, the results could obtained through the graphic only.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 25 0 100 200 300 400 0100 200 300 400 500 qe(mg/g) Cex 102(mg/L) Ap Langmuir Ap Freundlich Ap 0 100 200 300 400 0100 200 300 400 500 qe(mg/g) Cex 102 (mg/L) Av Langmuir Av Freundlich Av Figure 6 and 7 show the modelled curves for Langmuir and Freundlich, for each type of seaweed. Observing the graphics above it is possible to deduce that for Av both models adjust quite similarly, although at the end of the model lines one could prefer Langmuir’s Model since the curve adjusts slightly better to the final experimental points. Figure 6 – Ajdjusts for Av in the calcium adsorption isotherm assay. Figure 7 – Adjusts for Ap in the calcium adsorption isotherm assay.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 32 0 40 80 120 160 010 20 30 40 50 60 qe As (µg/g) CeAs (mg/L) Av ACa Table 4 presents Ca2+ and HOconcentrations in solution. This values were used to calculate the ionic product [Ca2+][HO-]2. The goal was to present a precipitation prediction for calcium hydroxide. These values are also much lower than Ca(OH)2 Ksp, which indicates that precipitation of the salt isn’t likely to occur. 3.4. Arsenic Adsorption Isotherm Figure 11 presents equilibrium biosorption data for arsenic removal by virgin seaweed and calcium-treated seaweed. The first thing to infer from figure 11 is that Av tend to remove much more arsenic from solution than ACa – 125 ± 5 µg/L and 10 ± 1 µg/L, respectively, at higher concentrations. Looking at figure 11 it is also possible to see that the concentrations of As in both algae tend to reach the maximum adsorption capacity in the solution concentration range of 15 - 20 mg/L. In ACa as the experimental points are so near each other, it is difficult to choose a value in which the monolayer coverage might be reached or to know if it actually exists in this assay. Other conclusion that is possible to take from this graphic, is that the results are not coherent with the previous ones, since this assay is made at a pH of approximately 6, which, according to figure 10, would mean that ACa should remove more arsenic than Av. Looking at the results of the titrations for pH 6, the potential surface charge for Av is Figure 11 - Experimental data obtained from the arsenic adsorption isotherm assay. Experimental conditions: V = 50 mL; m = 2 g; T = 25 C; t = 8 h; v = 180 rpm; pH 6 ± 0.5;
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 33 0 40 80 120 160 200 020 40 60 qeAs (µg/g) CeAs (mg/L) Av Av Freundlich Av Langmuir negative in that pH meaning that it is less probable that the anion (H2AsO4-) could bind to the algae surface by electrostatic attraction, in more quantity than in ACa, where the potential surface charge tends to be positive, therefore facilitating the binding of arsenic. However, the different pH adsorption assays were made using a mass ratio of 7 g/L while this isotherm used 40 g/L which can mean that a higher mass ratio for ACa can harm the arsenic adsorption when its concentration is higher than 15 mg/L. This isotherm was also adjusted to Langmuir and Freundlich models, previously explained in section 3.1, which originated the following results (Figure12). Considering a confidence level of 95%, the obtained parameters were not statistically significant for Freundlich model and for the Langmuir fitting on ACa. Table 5 presents the parameter values for the Langmuir model adjusted to Av.. Looking at figure 12, it is possible to see that, once again both models tend to adjust well to the experimental data. In figure 13, one can see that the models are very similarly adjusted, which means that both models adjust very similarly, making it impossible to draw any conclusions about the model that best describes arsenic adsorption behaviour in the ACa seaweed. It is possible to observe that for both Av and ACa, the adjusts present similar correlations so, as in the calcium adsorption isotherm, further statistical calculations had to be executed to best knowledge of the best fitting model. Figure 12 - Adjusts for Av in the arsenic adsorption isotherm assay.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 34 0 4 8 12 16 20 020 40 60 qe As(µg/g) CeAs (mg/L) ACa Aca Freundlich Aca Langmuir Table 5 presents the model adjustment parameters for Langmuir model and alga Av. The reason it was chosen to be presented this way is because the correlation coefficients for both models on the ACa arsenic adsorption isotherm and for Freundlich model also on alga Av were very low so, obtained parameters haven’t statistical significance Table 5 - Langmuir adjust parameters for Av (value ±interval for 95% confidence). Langmuir Model Qmax µg/g KL L/µg r2 Av 200 ± 86 0.03 ± 0.02 0.778 The adsorption capacities for both treated and non-treated algae demonstrated to be very low, which in a real situation would demand higher doses of the adsorbent that, as seen before can also lead to lower adsorption capacities. Comparing the Qmax value (200 ± 86 µg/g) of the virgin seaweed to the ones mentioned in table 1, particularly for As (V) it is possible to see its similarity with the agricultural residue – rice polish – that presented adsorption values of 147 µg/g [49] and the anaerobic biomass with 155 µg/g [50] of arsenic adsorption. We can also compare the value obtained in the present assay with the one obtained for the Lessonia nigrescens algae (a brown seaweed), 45.2 mg/g [55], and conclude that this example presents much better results Figure 13 - Adjusts for ACa in the arsenic adsorption isotherm assay
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 35 0.00 20.00 40.00 60.00 80.00 050 100 150 200 250 qe(µg/g) t (min) ACa than the studied species A. nodosum. Hansen et al. though, used initial As concentrations much higher (up to 600 mg/L) than the ones used in this project. This high concentration range are hardly found in real As-contaminated water and wastewater. In addiction, Tuzen et al. [54] reported considerable adsorption capacities (67.2 mg/g) for green algae, but for As(III) species. The literature values about arsenic adsorption in algae are, however, still scarce. 3.5. Arsenic Adsorption Kinetics In this section, the kinetics for ACa is presented (figure 14). By the observation of figure 14, we can see that the highest adsorption that’s represented is around 34 ± 15 µg/g (equilibrium value). The maximum adsorbed amount obtained here was not however according to the arsenic adsorption tests (section 3.3) at pH6. As referred previously, the mass ratio for this assay was different from the preliminary adsorption tests (40 g/L) which may be, once again, one factor of interference resulting in differences of results between assays. In order to clarify this results, a mass ratio adsorbent optimization assay should be conducted, since the project results suggest it as a highly important variable in the arsenic adsorption trough this algae type. Figure 14 - Experimental data obtained from the arsenic adsorption kinetics assay. Experimental conditions: V = 500 mL; m = 20 mg; t = 8 h; v = 200 rpm; pH 6 ± 0.5; CAs = 25 mg/L.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 36 It is also possible to see that in the first hour the adsorption tends to increase rapidly, stabilizing at that time, which can mean that this metalloid full adsorption is very quick and after about 1-2 hours the process is complete and the biosorption equilibrium was attained. Tuzen et al. [54] also reported 1 hour as the contact time for arsenic adsorption in U. cylindricum. Some authors such as Kim et al [71] which studies the arsenate adsorption in aluminum sludges, and Zhang et al [72] regarding the arsenate adsorption and desorption in soils, presented times of adsorption of 6 hours.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 37 4. Conclusions The present project was made on an exploratory basis, due to the lack of studies concerning arsenic biosorption on seaweed, and since the calcium treatment explored seems to have never been tested for this purpose, to the author knowledge. During the project some evaluations were made focusing the results obtained and some conclusions were possible to take. The calcium adsorption isotherm shown that maximum adsorption capacities of 261+-75 and 419+-384 mg/g (approx.) are reached when solutions of initial calcium concentrations are higher than 0.7 M. Similar behavior was observed for both Av and Ap algae. With the calcium treatment, calcium quantification in the algae increased to values around 35 mg/g. This value seems discrepant from the obtained in the previous assay, but they can be explained by the necessity to wash the algae after the calcium treatment in order to low its conductivity. The other conclusion taken by this assay is that it is not necessary to proceed to a protonation of this seaweed as the calcium adsorbed without it remains similar. Regarding the potentiometric titrations, the pHzc were determined for both Av and ACa presenting the values of 4.8 and 11.4 respectively. This means that the calcium treatment was responsible to ACa being positive in most of the pH range while the same didn’t happen with Av. This would mean that the calcium treated algae has more affinity with the arsenic anions present in solution, therefore being more able to retrieve them from the aqueous phase. The arsenic adsorption tests conducted at different pH’s concluded that pH 5 and pH6 are the preferential conditions to this process, and that for pH 6 (the pH used in most of the remaining assays) the adsorption presented for ACa rounded the 200 µg/g. However, adsorption isotherms, at pH 6 and using higher adsorbent dosages (40 g/L) showed much lower adsorbed amounts and also, a worst performance of ACa, compared to virgin form. The maximum adsorption capacity obtained in these conditions was 200 ± 86 µg/g (predicted by Langmuir modeling). Although this project results are still on a preliminary phase and are not very extensive, they show that brown seaweed tends to have a limited potential when it comes to As adsorption.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 38 5. Suggestions for Future Work Some suggestions for further work on this subject would be: Assays for the optimization of the mass ratio; Use different types of algae as green and red algae since they seem to present better results in arsenic removal; [53,54] Test for the increase of the organic material in the effluent after the treatment (TOC), along with a cost-benefit analysis to ensure that the removal of the exceeding organic material won’t become more expensive than the treatment with another commercial and more reliable adsorbent; Test for different temperatures, since the thermal amplitude is not the same in every country, and temperature can exert a significant influence; Instead of testing the biosorption, evaluate the ability of algae to bioaccumulation (with living algae), since the seaweed tend to have a metabolism that can naturally absorb arsenic, and maybe if exposed to a saturated environment the metabolism would adapt to the metalloid presence and metabolize it; Test this process for different metalloids such as selenium and antimony; although some chemical similarities exist, the biosorption of these elements can be more significant that usually behave as arsenic; Proceed to different low cost types of algae treatment that can improve its ability to remove arsenic (for example: a cationic surfactant)
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 39 6. References [1] Davis, T. A., Volesky, B., Mucci, A., 2003. A review of the biochemistry of heavy metal biosorption by Brown algae. Water Research 37, 4311-4330. [2] He, J., Chen, J. P., 2014. A comprehensive review on biosorption of heavy metals by algal biomass: Materials, performances, chemistry, and modelling simulation tools. Biosource Technology 160, 67-78. [3] Uluozlu, O. D., Sari, A., Tuzen, M., 2010. Biosorption of antimony from aqueous solution by lichen (Physcia tribacia) biomass. Chemical Engineering Journal 163, 382–388. [4] Garelick H, Jones H, Dybowska A, Valsami-Jones E., 2008. Arsenic pollution sources. Reviews of Environmental Contamination and Toxicology. 197:17-60. [5] Cunha, P. D. R., Duarte, A. A. L. S.. Remoção de Arsénio em águas para consumo humano. Escola de Engenharia, Universidade do Minho. [6] Smedley, P. L., Kinniburgh, D. G., 2002. A review of the source, behaviour and distribution of arsenic in natural waters. Applied Geochemistry 17, 517–568. [7] Masson, M., Schäfer, J., Blanc, G., Pierre, A., 2007. Seasonal variations and annual fluxes of arsenic in the Garonne, Dordogne and Isle Rivers, France. Science of the Total Environment 373, 196-207. [8] Seyler, P., Martin, J.-M., 1990. Distribution of arsenite and total dissolved arsenic in major French estuaries: dependence on biogeochemical processes and anthropogenic inputs. Marine Chemistry 29, 277–294. [9] Seyler, P., Martin, J.-M., 1991. Arsenic and selenium in a pristine river-estuarine system: the Krka, Yugoslavia. Marine Chemistry 34, 137–151. [10] Andreae, M.O., Andreae, T.W., 1989. Dissolved arsenic species in the Schelde estuary and watershed, Belgium. Estuarine, Coastal and Shelf Science. 29, 421–433. [11] Cullen, W.R., Reimer, K.J., 1989. Arsenic speciation in the environment. Chemical Reviews 89, 713– 764. [12] Edmunds, W.M., Cook, J.M., Kinniburgh, D.G., Miles, D.L., Trafford, J.M., 1989. Trace-element Occurrence in British Groundwaters. Res. Report SD/89/3, British Geological Survey, Keyworth. [13] Luo, Z.D., Zhang, Y.M., Ma, L., Zhang, G.Y., He, X., Wilson, R., Byrd, D.M., Griffiths, J.G., Lai, S., He, L., Grumski, K., Lamm, S.H., 1997. Chronic arsenicism and cancer in Inner Mongolia - consequences of
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 40 well-water arsenic levels greater than 50 mg l.1 . In: Abernathy, C.O., Calderon, R.L., Chappell, W.R. (Eds.), Arsenic Exposure and Health Effects. Chapman Hall, London, pp. 55–68. [14] Navarro, M., Sanchez, M., Lopez, H., Lopez, M.C., 1993. Arsenic contamination levels in waters, soils, and sludges in southeast spain. Bulletin of Environmental Contamination and Toxicology 50, 356–362. [15] Varsanyi, I., 1989. Arsenic in deep groundwater, in: D.L. Miles (Ed.), Proceedings of the 6th International Symposium on Water–Rock Interaction (WRI-6), Malvern, Publisher AA Balkema, Rotterdam/Brookfield 715–718. [16] Kurttio, P., Komulainen, H., Hakala, E., Kahelin, H., Pekkanen, J., 1998. Urinary excretion of arsenic species after exposure to arsenic present in drinking water. Archives of Environmental Contamination and Toxicology. 34, 297. [17] Silva, E. F., Zhang, C., Pinto, L. S., Patinha, C., Reis, P., 2004. Hazard assessment on arsenic and lead in soils of Castromil gold mining area, Portugal. Applied Geochemistry 19 (6), 887–898. [18] Cama, J., Rovira, M, Ávila, P, Pereira, M. R., Asta, M. P. Grandia, F., Martinez-Lladó, X., ÁlvarezAyuso, E., 2008. Distribución de arsénico en la región Ibérica. In: IBEROARSEN: distribución del arsénico en las regiones Ibérica e Iberoamericana. CYTED. [19] Garelick H, Jones H, Dybowska A, Valsami-Jones E., 2008. Arsenic pollution sources. Reviews of Environmental Contamination and Toxicology. 197:17-60. [20] Silva, A. F., Fiúza, A. 2011. Distribuição e comportamento do arsénio em ambientes naturais. Faculdade de Engenharia, Universidade do Porto. [21] Ungureanu, G., Santos, S., Boaventura, R., Botelho, C., 2015. Arsenic and Antimony in water and wastewater: overview of removal techniques with special reference to latest advances in adsorption. Journal of Environmental Management 151, 326–342. [22] Fowler, B. A., Selene, C. H., Robert, J. C., Dexter, L. J., Sullivan Jr, W., Chen, C. J., 2015. Handbook on the Toxicology of Metals. Chapter 28 – Arsenic. Fourth Edition. Academic Press 581–624. [23] Hughes, M., 2002. Arsenic toxicity and potential mechanisms of action. Toxicology Letters 133, 1–16. [24] Mandal, B. K., Suzuki, K. T., 2002. Arsenic round the world: a review. Talanta 58, 201-235. [25] EPA – United States Environmental Protection Agency. Arsenic Compounds. (www.epa.gov/ttnatw01/hlthef/arsenic.html) – accessed at 19th October 2014 [26] Olsen, J. L., Zechman, F. W., Hoarau, G., Coyer, J. A., Stam, W. T., Valero, M., Aberg, P., 2010. The phylogeographic architecture of the fucoid seaweed Ascophyllum nodosum: an intertidal ‘marine tree’ and survivor of more than one glacial–interglacial cycle. Journal of Biogeography 37, 842-856.
Study of arsenic biosorption in Ascophyllum nodosum Environmental Engineering Master Thesis 41 [27] Barsanti, L., Gualtieri, P., 2014. Algae Anatomy, Biochemistry, and Biotechnology. Second Edition. CRC Press. 267-308. [28] Freitas, O., Boaventura, R., Delerue-Matos, C., 2006. Adsorption study of lead by Ascophyllum nodosum using a factorial experimental design. Combined and Hybrid Adsorbents – NATO Security through Science Series, 269-274. [29] Somasundaran, P., 2004. Encyclopedia of surface and Colloid Science. Volume 5. Marcel Dekker. [30] Ungureanu, G., 2013. Arsenic and Antimony – an overview. Unpublished. [31] Mondal, P., Bhowmick, S., Chatterjee, D., Figoli, A., Bruggen, B., 2013. Remediation of inorganic arsenic in groundwater for safe water supply: A critical assessment of technological solutions. Chemosphere 92, 157–170. [32] Thakur, L. S., Semil, P., 2013. Removal of arsenic in aqueous solution by low cost adsorbent - A short review. International Journal of ChemTech Research 5 (3), 1299-1308. [33] Han, B., Runnells, T., Zimbron, J., Wickramasingh, R., 2002. Arsenic removal from drinking water by flocculation and microfiltration. Desalination 145 (1–3), 293–298. [34] Mohan, D., Pittman Jr., C. U., 2007. Arsenic removal from water/wastewater using adsorbents – a critical review. Journal of Hazardous Materials 142, 1-53. [35] Jeppu, G. P., Clement, T. P., 2010. A modified Langmuir-Freundlich isotherm model for simulating pHdependent adsorption effects. Journal of Contaminant Hydrology 129–130, 46–53. [36] Faculdade de Engenharia, Universidade do Porto. Permuta Iónica. (paginas.fe.up.pt/~mgi00015/Ensdist/versao_web/teoria/teoria_base.htm) accessed on 18th October 2014 [37] Sousa, A. C., 2014. Remoção de arsénio e antimónio de soluções aquosas por biossorção. Master’s Thesis, Faculdade de Engenharia, Universidade do Porto. [38] Tamburini, J. R., Liang, H. C., Billin, S. J., 2010. Single Process Arsenic and Antimony Removal Using Coagulation and Microfiltration. Tetra Tech, Colorado State University. Tailings and Mine Waste 2010. [39] Mohan, D. and Pittman, C.U., 2007. Arsenic removal from water/wastewater using adsorbents – A critical review. Journal of Hazardous Materials 142 (1-2), 1-53. [40] Petersen, F., Aldrich, C., Esau, A., Qi, B. C., 2005. Biosorption of heavy metals from aqueous solutions. Report to the Water Research Commission. [41] Stadie, N. P., 2013. Synthesis and thermodynamic studies of physisorptive energy storage materials. Dissertation (Ph.D.), California Institute of Technology.