Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electrochemical characterization of carminic acid towards the use as an electrochemical molecular beacon… Gorbunova, Santa Maria 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_may_gorbunova_santamaria.pdf [ 1.79MB ]
Metadata
JSON: 24-1.0167174.json
JSON-LD: 24-1.0167174-ld.json
RDF/XML (Pretty): 24-1.0167174-rdf.xml
RDF/JSON: 24-1.0167174-rdf.json
Turtle: 24-1.0167174-turtle.txt
N-Triples: 24-1.0167174-rdf-ntriples.txt
Original Record: 24-1.0167174-source.json
Full Text
24-1.0167174-fulltext.txt
Citation
24-1.0167174.ris

Full Text

ELECTROCHEMICALCHARACTERIZATION OF CARMINICACID TOWARDS THE USE AS ANELECTROCHEMICAL MOLECULARBEACON FOR NUCLEIC ACIDDETECTIONbySanta Maria GorbunovaB.Sc., University of Brighton, 2011A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinThe Faculty of Graduate and Postdoctoral Studies(Chemistry)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)April 2015© Santa Maria Gorbunova 2015AbstractWorldwide, more than a million people die from tuberculosis (TB) every year. Although the diseaseis curable, treatment is complicated by multi-drug resistant and extensively drug-resistant TB strains.To detect TB and differentiate between its strains, a sensitive and specific point-of-care device is re-quired. Previous studies show that carminic acid (CA), an anthraquinone derivative, is suitable as anelectrochemical molecular beacon due to the ability to switch on and off its electrochemical activity onits dimerization.Characterization of the electrochemical activity of CA at low concentrations (1 µM to 1 mM) over arange of pH values was performed using methods such as cyclic voltammetry, square wave voltammetryand Koutecky´-Levich analysis on a rotating disk electrode. CA species of different protonation, which arepredominant at pH 1.1, pH 4.1, pH 6.6 and pH 10.5, were examined in more detail. All measurementswere carried out on a glassy carbon electrode in phosphate buffer solution electrolyte.It was found that CA undergoes a diffusion limited two proton two electron redox reaction with anoverall peak potential shift of 61 mV per pH unit. Electrochemical measurements of the fully protonatedCA resulted in additional current peaks that were assigned to an adsorption process of a CA reductionproduct. Generally, CA has faster electron transfer kinetics in more acidic environment and no elec-trochemical activity was observed for the fully deprotonated CA species at pH 10.5. While SWV couldbe used for quantitative analysis of CA for the concentrations up to 1 mM, its redox current signal wasdetermined not to be concentration dependent at high measurement frequencies. These frequenciescan also be adjusted to be more sensitive towards either the redox peak potentials with sharper peaksat low frequencies or the electron transfer kinetics based on kinetic dependent peak currents at highfrequencies. The limit of detection for CA at pH 7.0 was found to be as low as 10 nM when measuredusing 200 Hz SWV.iiPrefaceThe experimental work presented in this thesis is an original and unpublished work by the author, S.M. Gorbunova, and the development of the experimental procedures and data analyses were done incollaboration with the supervisor, D. Bizzotto.iiiTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiNomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Tuberculosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Carminic acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.3 Scope of the project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Electrochemistry Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373.2 Electrochemical cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383.3 Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403.4 Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40ivTable of Contents4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414.1 Carminic acid electrochemistry at different pH . . . . . . . . . . . . . . . . . . . . . . . . . 414.2 Characterization of carminic acid electrochemical activity at low pH . . . . . . . . . . . . 484.3 Study of the electron transfer kinetics using an RDE . . . . . . . . . . . . . . . . . . . . . 505 Summary and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61AppendicesA Kinetic Analysis of Cobalt Hexammine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64vList of Tables3.1 Composition of electrolyte at the experimental pH values. . . . . . . . . . . . . . . . . . . 383.2 List of the experimental values of CV sweep rates, SWV measurement frequencies andRDE rotations rates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404.1 CA redox peak potentials and the estimated formal electrode potential (E0’) in different pHenvironments obtained from CV in 100 µM CA using a GC WE, sweep rate – 20 mV/s. 424.2 Summarized results of 100 µM CA kinetic parameters on a GC electrode. . . . . . . . . 54A.1 Data obtained from cobalt hexammine CV . . . . . . . . . . . . . . . . . . . . . . . . . . . 65A.2 Summarized cobalt hexammine kinetic parameters obtained by different methods. . . . 68viList of Figures1.1 Carminic acid structure with indicated deprotonation sites. . . . . . . . . . . . . . . . . . . 41.2 Fractions of CA species at different pH values. . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Classic polarography of 0.86 mM CA in 0.1 M LiClO4, scan rate: 4 mV/s. . . . . . . . . 51.4 SWV of CA modified graphite-polyester composite electrode in organic and aqueous so-lutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.5 (a) Two electron two proton, (b) two electron one proton and (c) two electron reduction ofquinone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61.6 Half-wave potential and pH relationship of 2-hydroxy-1,4-benzoquinone. . . . . . . . . . 71.7 A comparison of a 3 proton 2 electron and 2 proton 2 electron redox reaction of a hydrox-ylquinone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71.8 Two step quinone reduction steps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.9 Pourbaix diagram for the disproportionation redox reaction of 2-AQMS. . . . . . . . . . . 91.10 Scheme of squares diagram of major quinone states. . . . . . . . . . . . . . . . . . . . . 101.11 Scheme of squares fitted for anthraquinones . . . . . . . . . . . . . . . . . . . . . . . . . . 101.12 A redox mechanism of a quinone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.13 CV of 2 mM 1,4-benzoquinone in DMSO with 0.2 M Bu4NPF6 at scan rate of 100 mV/swithout addition of benzoic acid, with 0.03 M and 1 M benzoic acid. . . . . . . . . . . . . 121.14 CV of 1 mM anthraquinone-2-sulfonic acid at scan rate of 100 mV/s on a gold electrodein aqueous buffered solutions of different pH. . . . . . . . . . . . . . . . . . . . . . . . . . 121.15 (a) Bound CA concentration dependence on DNA concentration. (b) Emitted fluorescencespectra of CA and CA - DNA complex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141.16 Mechanism of CA as ECMB in a hairpin loop form. . . . . . . . . . . . . . . . . . . . . . . 151.17 Mechanism of CA as ECMB in a stem loop form. . . . . . . . . . . . . . . . . . . . . . . . 151.18 CV of CA as an ECMB in its dimer form, as ECMB in its monomer form and in its freeform (CA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16viiList of Figures1.19 Differential pulse voltammetry of CA-ECMB with a different complementarity nucleic acidsequences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.1 Schematic representation of electrical double layer. . . . . . . . . . . . . . . . . . . . . . . 202.2 Potential drop across the electrical double layer. . . . . . . . . . . . . . . . . . . . . . . . . 212.3 Energy barrier symmetry for different α values. . . . . . . . . . . . . . . . . . . . . . . . . 232.4 Example of a Tafel plot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242.5 Schematic representation of a cell as a circuit. . . . . . . . . . . . . . . . . . . . . . . . . . 262.6 The potential applied in linear sweep and the resulting charging current. . . . . . . . . . 272.7 Applied potential and the resulting CV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282.8 Potential steps in staircase CV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302.9 Potential steps in DPV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302.10 Current response to the applied potential steps. . . . . . . . . . . . . . . . . . . . . . . . . 312.11 Example of DPV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312.12 Potential profile in SWV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322.13 Example of SWV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332.14 Mass transfer velocities and flows at the RDE. . . . . . . . . . . . . . . . . . . . . . . . . . 342.15 Example of current vs. potential measurement using RDE. . . . . . . . . . . . . . . . . . 352.16 Example of a Koutecky´-Levich plot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363.1 Electrochemical cell setup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394.1 CV of 100 µM CA in different pH phosphate buffer solutions on GC WE (area 7.0 mm2),sweep rate of 20 mV/s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424.2 Peak potential and pH relationship of 100 µMCA in different pH phosphate buffer solutionson GC at a sweep rate of 20 mV/s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434.3 CV of 100 µM CA in pH 6.6 phosphate buffer at different sweep rates on GC WE (area7.0 mm2). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444.4 Peak current and pH relationship of 100 µM CA at different sweep rates on GC. . . . . 444.5 Relationship between current and square root of sweep rate . . . . . . . . . . . . . . . . 454.6 CV of different concentration CA in pH 6.6 phosphate buffer on GC (area 7.0 mm2) witha sweep rate of 20 mV/s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464.7 Relationship between peak current (log) and concentration of CA (log) . . . . . . . . . . 46viiiList of Figures4.8 Proposed CA redox mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474.9 First cycles of 1 µM CA CV in a pH 1.0 solution on GC, sweep rate 10 mV/s. . . . . . . 484.10 1 µM and 100 µM CA in pH 1.0 phosphate buffer on GC at different potential ranges,covering only the more negative redox peaks of interest or only the more positive redoxpeaks potential range. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494.11 The change in ratio of the CA redox peaks in CV at pH 1.0 on GC, sweep rate 20 mV/s. 504.12 Relationship between current and square root of sweep rate for the additional peak at pH1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504.13 RDE reduction curves of 100 µM CA at pH 1.1, pH 4.1, pH 6.7 and pH 10.5, all in phos-phate buffer solution on a GC (area 7.0 mm2). . . . . . . . . . . . . . . . . . . . . . . . . . 514.14 Koutecky´-Levich plot for 100 µM CA at pH 1.1, pH 4.1 and pH 6.7 . . . . . . . . . . . . . 524.15 Tafel plots for CA at pH 1.1, pH 4.1 and pH 6.7 with fitted Tafel slopes. . . . . . . . . . . 534.16 SWV of 100 µM CA at different pH at 2 Hz and at 200 Hz frequencies . . . . . . . . . . . 554.17 Relationship between δi (log) and frequency (log) of 100 µM CA at different pH (SWV). 564.18 Relationship between current (log) and frequency (log) of several CA concentrations atdifferent pH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574.19 LOD for CA at pH 7.0 at 2 Hz and 200 Hz. . . . . . . . . . . . . . . . . . . . . . . . . . . . 58A.1 CV of cobalt hexammine at different sweeprates . . . . . . . . . . . . . . . . . . . . . . . . 65A.2 Relationship of peak current and square root of sweep rate of cobalt hexammine . . . . 66A.3 Cobalt hexammine RDE reduction curves at 500 to 3000 rpm. . . . . . . . . . . . . . . . 66A.4 Koutecky´-Levich plot for cobalt hexammine. . . . . . . . . . . . . . . . . . . . . . . . . . . 67A.5 Tafel plot for cobalt hexammine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68ixNomenclature2-AQMS Anthraquinone-2-sulfonic acid, page 72,6-AQDS Anthraquinone-2,6-disulfonic acid, page 7AIDS Acquired immune deficiency syndrome, page 1AQ Anthraquinone, page 7C* Electrolyte bulk concentration, page 20CA Carminic acid, page 3C Capacitance, page 19C Cytosine, page 13Cd Capacitance of the double layer, page 26CE Counter electrode, page 25CJ Concentration of species J, page 22CSCE Capacitance of SCE, page 26CV Cyclic voltammetry, page 15DI Deionized, page 37DNA Deoxyribonucleic acid, page 13DPV Differential pulse voltammetry, page 16dsDNA Double stranded DNA, page 13E0′Formal electrode potential, page 22ECMB Electrochemical molecular beacon, page 14xNomenclaturee Elementary charge, page 20Ei Initial potential, page 32Ep/2 Half peak potential, page 28Ep Peak potential, page 28ƒ Rotation frequency in RDE, page 33F Faraday constant, 96 485 s.A.mol-1, page 22GC Glassy carbon, page 38G Guanine, page 13HIV Human immunodeficiency virus, page 1i0 Exchange current, page 23δi Current difference, page 31IHP Inner Helmholtz plane, page 19iK Kinetically limited current, page 35il Steady-state current, page 34ipa Anodic peak current, page 22ipc Cathodic peak current, page 22k0 Standard rate constant, page 23kB Boltzmann constant, 1.38 × 10−23 m2 .kg.s-2 .K-1, page 20kb Reverse rate constant, page 22kƒ Forward rate constant, page 22LOD Limit of detection, page 39MB Molecular beacon, page 2MDR Multi-drug resistant, page 1xiNomenclaturen Number of electrons, page 22OHP Outer Helmholtz plane, page 19ppb Parts per billion, page 37QH2 Hydroquinone, page 8QH• Quinone radical, page 8qM Charge excess on the electrode surface, page 19Q Quinone, page 8q Charge, page 19qS Charge excess in the solution, page 19RDE Rotating disk electrode, page 33E Potential, page 19redox Reduction and oxidation, page 21RE Reference electrode, page 25rpm Revolutions per minute, page 40R Gas constant, 8.314 J.mol-1.K-1, page 22RS Resistance of solution, page 26SCE Saturated calomel electrode, page 11ssDNA Single stranded DNA, page 13SWV Square wave voltammetry, page 4TB Tuberculosis, page 1TST Tuberculin skin test, page 2T Temperature, page 20t Time, page 29xiiNomenclatureWE Working electrode, page 25XDR extensively drug-resistant, page 1z Ionic charge, page 20α Transfer coefficient, page 23δO Diffusion layer thickness, page 34ϵ0 Dielectric permittivity in the vacuum, page 20ϵr Relative dielectric permittivity of the electrolyte, page 20η Overpotential, page 24κ−1 Debye-Hu¨ckel length, page 19ν Kinematic viscosity, page 34τ Time fraction, page 29υ Sweep rate, page 26ω Angular velocity, page 33xiiiAcknowledgementsI would like to thank everyone who has supported me in this period of my life. I would not have done itwithout your help.First of all, I would like to show appreciation to my supervisor, Dr. Dan Bizzotto. Thank you forwelcoming me to your lab and providing the valuable guidance and support throughout my stay here.The current and past Bizzotto group members, you have made me feel warm and cozy at my sec-ond home - the lab. I appreciate your company and I have truly enjoyed the numerous discussions,sleepovers and outdoor activities we have had.The UBC Rueda club members, you have helped me get back on my feet at my toughest times anddance the problems away. I have learned so much from you. Thank you for your companionship and allthe energy and inspiration you gave me.My family and the closer friends, I am deeply grateful for always having you by my side regardless ofcircumstances. Your infinite love and support have kept me going and have made my dreams possible.Finally, I would also like to thank the glassblower, Brian Ditchburn, for the amazing glassware hemakes; Dr. Carl F. Perez from Metaara Medical Technologies Inc. for bringing this project to our atten-tion and being part of the team; Natural Sciences and Engineering Council of Canada (NSERC) andCanadian Institutes of Health Research (CIHR) for the funding; and the University of British Columbia,specifically, Advanced Materials and Process Engineering Laboratory (AMPEL) for providing the facili-ties and the Department of Chemistry for giving me the opportunity to pursue the graduate studies.xivChapter 1Introduction1.1 TuberculosisTuberculosis (TB) infection is caused by Mycobacterium tuberculosis bacteria [1, 2] resulting in fever,night sweats, weight loss and other symptoms, amongst which cough and sneezing help transmit thedisease[1]. As symptomsmay be not noticeable for months, diagnosis and treatment is delayed, spread-ing TB further. TB is the second most deadly infectious disease worldwide, with numbers of lethal casesexceeded only by HIV/AIDS, and it is the number one cause of death for HIV infected people. About 30%of the world population are latent TB carriers, 10% of which will develop the disease, and this numberincreases for people with weakened immune system. Multi-drug resistant (MDR) TB strains complicatethe treatment due to their resistance to the most commonly used first-line antibiotics like rifampicin andisoniazid, whereas extensively drug-resistant (XDR) strains are not susceptible to even larger numberof the second-line antibiotics. Improper treatment not only wastes time and money but also increasesTB resistivity to antibiotics. In 2012, 8.6 million people were diagnosed with TB and 1.3 million patientsdied from it, the vast majority of which were from developing countries with restricted access to hospitaland/or diagnostic facilities. Almost half a million of TB cases were due to MDR-TB, every tenth of whichwere XDR; these strains were found to be present in every country where TB is diagnosed. [1]The same year in Canada, 0.0047% to 0.0048% of population were diagnosed with TB [3], which isone of the lowest numbers in the world [4]. 86% of these cases have been fully treated, 10% resultedin death and the remaining are still being in care. [3] The majority of TB cases involved foreign citizens[3]; however, among the Canadian-born population, Aboriginal (especially, First Nation and Inuit) peo-ple were responsible for 64% to 83% of the rest of the cases depending on the province [3, 4]. Thispopulation group is at higher risk to develop TB disease due to immune system commonly weakenedby HIV/AIDS, diabetes and/or other illnesses and due to inaccessibility of diagnostic facilities. HealthCanada stresses the necessity of preventing and controlling TB infection among the risk groups. [4]It is therefore important to develop a cost effective, sensitive and selective diagnostic tool that would11.1. Tuberculosisenable a quick and accurate detection and identification of TB strains not only in a hospital environmentbut also in rural regions with limited access to current diagnostic facilities, allowing immediate start of thetreatment with appropriate anti-TB drugs, decreasing number of lethal cases and spread of the infection.1.1.1 Current TB diagnosticsOne of the commonly used tests for TB detection in high risk areas is a tuberculin skin test (TST). It isbased on a delayed immunological response of the organism to the injected tuberculin and the resultcan be measured at the peak response 48 to 72 hours after the injection. [5] Although TST has anadvantage of detecting latent TB, it is not sensitive and specific enough for active TB confirmation. [6]Chest radiograph (chest X-ray) detects abnormalities, like lesions, in lungs. Just like in case of TST,additional tests are required to confirm TB disease; however, in case of a positive TST, chest radiographcan exclude pulmonary TB if no abnormalities are detected. [7] Sputum smear microscopy detects acid-fast bacilli in sputum samples. [5] It requires a trained personnel as each slide has to be examinedmanually. [6] False positive results are also not uncommon in this technique. [5] From the typical TBdiagnostic tests, only an incubated culture that is identified to be TB is considered to be a confirmationof the infection [5]; however, this test might take weeks to complete [6].Although these tests have been used for a long time, they do not provide enough sensitivity, specificityor flexibility to detect and confirm TB infection in a timely manner. One of the newest TB diagnostictools, Xpert® MTB/RIF, overcomes at least some of these issues. It consists of single-use cartridges,containing unprocessed test samples, and a multi-module platform that can accommodate from one upto 48 cartridges at a time. The platform is fully automated and, through isolation of nucleic acids andpolymerase chain reaction amplification, it detects TB in less than 2 hours. [2] The detection processemploys molecular beacon (MB) [2] probes which allow direct and continuous measurement in a solution[8]. For simultaneous differentiation between drug-susceptible and drug-resistant strains, an 81-base-pair core region sequence of the bacterial RNA polymerase β subunit is also included with an additionalMB probe as the most mutations for drug resistance occur in this part of the sequence. Xpert® MTB/RIFhas proved to be a sensitive and specific tool for TB detection and differentiation of the drug-resistantstrains. It also detects both, live and dead, TB bacteria. Although the results presented by Xpert®MTB/RIF are very satisfactory, it can be quite expensive, requires trained laboratory staff to run theinstrument and could be problematic to use in underdeveloped and resource poor areas. [2]To develop a point-of-care device for TB detection and differentiation between its strains, a similar21.2. Carminic acidMB method in combination with electrochemical detection is being considered. Previous studies haveshown that carminic acid (CA) exhibits properties that make it a suitable probe for this application whichwill be described later in this chapter. [8, 9, 10]1.2 Carminic acidCarminic acid (CA) (also known as natural red 4, E120, CI 75470, CAS 1260-17-9) is a red colourednatural dye which is used in many food, textile and cosmetics products. [11, 12] It is obtained fromparasitic cochineal insects (Dactylopius coccus) which live on Opuntia type cacti [11, 12, 13]. Driedinsects are ground [12, 14] and dye is then extracted with alkaline aqueous solution [12] yielding around10% to 22% of insect original dry weight [15, 16]. Cochineal dye is available in several forms (identitiesof which are sometimes mislabeled). [12] First is a crude cochineal extract which often contains not onlyCA but also insect body parts [12, 13, 14], biological contaminants [12, 14] and ionic salts from extractionprocess [14]. Another is carmine, derived from the extract, which is a CA and aluminum complex as acalcium salt [12, 13] or in its free form [12]. CA is obtained by purifying the cochineal extract, andaminocarminic acid (also called acid-stable carmine) is formed by reacting CA with ammonia. [12]1.2.1 Properties of carminic acidCA (7-a-D-glycopyranosyl)-9,10-dihydro-3,5,6,8-tetrahydroxyl-methyl-9,10-dioxo-2-anthracenecarboxylicacid [17]) (Figure 1.1) is an anthraquinone with an attached hexose sugar, a carboxylic acid, a methyland four hydroxyl groups [12]. Due to the number of oxygen atoms and hydroxyl groups, it forms hy-drogen bonds, making it soluble in water. [12, 18] The sugar group is said to not participate in thecharge delocalization of the chromaphore and, hence, to not be involved in the optical properties of themolecule.[17] CA is stated to be a triprotic (or possibly a tetraprotic) acid with confirmed pKa values of2.81-3.42, 5.43-6.20, 8.10-8.94 [11, 19, 20, 21] and a probable estimated pKa value of 13. [20] Fullyprotonated CA form (H4CA) is dominant at pH 2, H3CA- at pH 4, H2CA2- at pH 7 and HCA3- at pH 9 (thedistribution diagram is given in Figure 1.2). Since the pKa of the fourth proton is quite high, only a smallfraction of CA4- might be detected at pH 10. [20] Also, just like other anthraquinone molecules, CA hasa tendency to dimerize due to hydrogen bonding or pi-pi electron stacking [9, 22] which is assumed tolead to loss of its electrochemical activity [9, 8, 10].31.2. Carminic acidFigure 1.1: Carminic acid structure with indicated deprotonation sites at pKa1= 2.81, pKa2= 5.43 andpKa3= 8.10. Reprinted with permission from [20]. Copyright (2015) American Chemical Society.Figure 1.2: Fractions of CA species at different pH values. Reprinted with permission from [20]. Copy-right (2015) American Chemical Society.The electrochemical activity of CA has been studied to some extent on mercury and carbon elec-trodes. Polarographic experiments (Figure 1.3) show two reduction curves - one of which follows aquasi-Nernstian behavior and the other one a non-Nernstian behavior; however, it is believed that theCA sample used in the experiment had impurities which might have affected the results. [11] Squarewave voltammetry (SWV) experiments (with selected optimum parameters: potential step 4 mV, ampli-tude 25 mV, frequency 15 Hz) show different results depending on electrolyte solution (Figure 1.4): twowell defined peaks at -1.555 V and -1.185 V and a small shoulder at around +0.6 V in organic solvent(0.1 M Bu4NPF6 in acetonitrile) and a single peak at -0.640 V with numerous not well defined signalsin the range of around -1.7 V to +0.5 V in aqueous solution (0.25 M sodium acetate and 0.25 M aceticacid aqueous buffer). [23]41.2. Carminic acidFigure 1.3: Classic polarography of 0.86 mM CA in 0.1 M LiClO4, scan rate: 4 mV/s. Reprinted withpermission from [11].Figure 1.4: SWV of CA modified graphite-polyester composite electrode in organic (on the left) andaqueous (on the right) solutions. Step potential 4 mV, amplitude 25 mV and frequency 15 Hz. Adaptedfrom [23] (original Figures 3 and 5) with kind permission from Springer Science and Business Media.51.2. Carminic acid1.2.2 QuinonesAlthough, there are notmany relevant electrochemical measurements of CA, properties of other quinones(especially, anthraquinones) can be used as a reference due to the structural similarities with CA. The re-dox reaction of anthraquinones is pH dependent. [24] It is known that quinones undergo two electron twoproton reduction [24, 18, 25] (Figure 1.5 a) at acidic pH, either two electron one proton reduction[24, 18](Figure 1.5 b) or two electron reduction with no protons involved (Figure 1.5 c) at neutral pH and twoelectron reduction with no protons involved at basic pH [24]. The peak current decreases in solutionswith higher pH values, and the change of reduction peak potential with pH, generally, follows Nernstianbehavior shifting by about 60 mV per pH unit [18]; however, the slope of the peak potential and pH rela-tionship for quinones containing a hydroxyl group has been observed to vary (Figure 1.6) [26]. Typically,the slope is 60 mV/pH unit, which corresponds to 2 electron 2 proton process, but between the two pKavalues of a hydroquinone (for example, pKa1 = 4.2 and pKa2 = 8.7 for 2-hydroxy-1,4-benzoquinone andpKa1 = 4.0 and pKa2 = 9.0 for 2-hydroxy-1,4-naphthoquinone) the slope changes to about 90 mV/pHunit, which suggests a 3 proton 2 electron process. This additional proton is explained to be due toprotonation of the deprotonated hydroxyl group at these pH values (see, Figure 1.7). Quinones havealso been found to start decomposing at pH 9 and lose their ability to be electrochemically reduced atmore basic conditions. [26]Figure 1.5: (a) Two electron two proton, (b) two electron one proton and (c) two electron reduction ofquinone. Reprinted from [18], copyright (2011) Partha Sarathi Guin et al.61.2. Carminic acidFigure 1.6: Half-wave potential and pH relationship of 2-hydroxy-1,4-benzoquinone. Reprinted from[26], Copyright (2015), with permission from Elsevier.Figure 1.7: A comparison of a (a) 3 proton 2 electron and (b) 2 proton 2 electron redox reaction of ahydroxylquinone. Reprinted from [26], Copyright (2015), with permission from Elsevier.Diffusion coefficients measured using mercury coated platinum microelectrodes of, for example, an-thraquinone (AQ), anthraquinone-2-sulfonic acid (2-AQMS) and anthraquinone-2,6-disulfonic acid (2,6-AQDS) at diffusion limited conditions are, respectively, 7 ×10-6 to 1 ×10-5 cm2/s, [27, 28] 2 ×10-6 to71.2. Carminic acid7 ×10-6 cm2/s and 3 ×10-6 to 6 ×10-6 cm2/s. [27] It shows a linear or close to linear relationship be-tween square root of sweep rate and peak current, and there is no or little adsorption observed on aglassy carbon electrode surface [18]. This kind of quinone-hydroquinone reduction peak is stated tobe either generally reversible [7, 18, 29] due to resonance stabilization of quinone radical intermediate(and irreversible only if quinone undergoes side reactions) [29] or reversible only in aprotic medium andnon-reversible in protic medium [30]. Some quinones are shown to undergo a quasi-reversible redoxreaction [25, 31]. The difference in quinone electrochemistry in different pH solutions is explained byhydrogen bonding stabilizing the intermediate form in acidic media. [24] Quinone (Q) reduction to hy-droquinone (QH2) can be generally expressed as Q + 2H+ + 2e- Š QH2 [18, 32] and this reaction hasshown to involve a formation of a quinone radical (QH•) intermediate (Figure 1.8) [24, 27, 30, 32, 33, 34].Due to radical instability, one of the quinone radicals readily undergoes a reduction (QH• to QH2) whileanother one - an oxidation (QH• to Q), this disproportionation reaction resulting in a Q and QH2 redoxcouple. The possible quinone-hydroquinone redox forms, including radical intermediates, for 2-AQMSin different pH media is presented in Pourbaix diagram (Figure 1.9). [27, 32]Figure 1.8: Two step quinone reduction steps. Reproduced from [30] with permission of The RoyalSociety of Chemistry.81.2. Carminic acidFigure 1.9: Pourbaix diagram for the disproportionation redox reaction of 2-AQMS. The line at the mostpositive potential represents a one electron redox reaction between QH• and QH2 and the line at themost negative potential - a one electron redox reaction between Q and QH•. Due to radical instability,these reactions often cannot be observed, resulting in a two electron redox reaction between the Qand QH2 redox couple which is represented by the the line that forms in the middle. Reprinted withpermission from [27]. Copyright (2015) American Chemical Society.The number of reduction reaction steps is not completely clear and has been stated to be either asingle step reduction [18] or a two step reduction involving first a rate limiting, energetically unfavorable,’quinone to quinone radical’ step and second a very quick energetically favorable, ’quinone radical tohydroquinone’ step (Figure 1.8) [30, 33] with a 9 mV difference in potential between these two reductionsteps [27]. However, this mechanism could be different at extreme pH values and a square schemediagram (Figure 1.10) can be used to represent the possible step pathways [30, 33]. Assuming nodimerization occurs [25], for two electron two proton reduction, this 9 member diagram has six pKa andsix potential values, where the pKa values for hydroquinones are ordered as follows: pKa4 < pKa1 <pKa5< pKa2 < pKa6 < pKa3 [34]. Figure 1.11 depicts a square scheme diagram fitted for 2,6-AQDS and2-AQMS showing quinone reduction mechanisms in different pH media. At strongly acidic pH, thesecompounds undergo proton-electron-proton-electron transfer, at slightly acidic to neutral pH - electron-proton-electron-proton transfer pathway, at slightly basic pH - electron-electron-proton-proton transfer[24] and at strongly basic pH - electron-electron transfer pathway [24, 7]. The fit is based on the peakcurrent and peak potential of the redox reaction at these pH values and it is assumed that there might beother mechanisms or that several mechanisms might be valid at the same time. [24] A quinone redox91.2. Carminic acidmechanism is obtained by combining the information from Figure 1.5 and Figure 1.10. The resultingscheme (Figure 1.12) shows the different products that can be acquired from a quinone electrochemicalreduction at different pH.Figure 1.10: Scheme of squares diagram of major quinone states. Reprinted with permission from [24].Copyright (2015) American Chemical Society.Figure 1.11: Scheme of squares fitted for anthraquinone-2,6-sulfonic acid and anthraquinone-2-sulfonicacid where horizontal lines show the electron transfer and vertical lines the proton transfer. Reprintedwith permission from [24]. Copyright (2015) American Chemical Society.101.2. Carminic acidFigure 1.12: A redox mechanism of a quinone.Generally, two sets of redox peaks are observed in hydroxy-9,10-anthraquinone electrochemistry:one reversible or quasi-reversible redox peak pair around +500 mV with respect to saturated calomelelectrode (SCE) and another irreversible reduction peak at around -600 mV to SCE which also followsthe Nernstian behavior, shifting by 50-60 mV per pH unit, and shows some reversibility in more acidicsolutions. [18] Similarities can also be observed with 1,4-benzoquinone (Figure ??). [28] The reasonfor having two sets of redox peaks is assumed to be either a hydrogen evolution from the hydroxylgroups neighboring the quinone [18] or, in addition to a diffusion peak, another peak might form due toadsorption process on the electrode [31]. Compared to unsubstituted anthraquinones, the peak potentialshift to more positive potentials has been observed for anthraquinones that have hydroxyl groups in theirstructures. [22] Changes in anthraquinone electrochemistry with pH, for the redox peaks at the morenegative potential, can be seen in Figure 1.14. [24]111.2. Carminic acidFigure 1.13: CV of 2 mM 1,4-benzoquinone in DMSO with 0.2 M Bu4NPF6 at scan rate of 100 mV/swithout addition of benzoic acid (solid line), with 0.03 M (dotted line) and 1 M (dashed line) benzoic acid.Reprinted from [28], Copyright (2015), with permission from Elsevier.Figure 1.14: CV of 1 mM anthraquinone-2-sulfonic acid at scan rate of 100 mV/s on a gold electrodein aqueous buffered solutions of different pH. Reprinted with permission from [24]. Copyright (2015)American Chemical Society.121.2. Carminic acid1.2.3 Interaction with DNAMolecules like CA (dihydroxyanthraquinones with at least one sugar moiety) are said to form complexeswith deoxyribonucleic acid (DNA) through introduction of the sugar into the DNA grooves and inter-calating the anthraquinone between cytosine (C) and guanine (G) base pairs. [35] Due to the planarstructure, anthraquinones are known to intercalate DNA even if they do not have a sugar group attached.[24, 36] However, anthraquinone and DNA binding studies with ethidium bromide as a fluorescent probestrongly suggest that this interaction is more likely to be groove binding than intercalating. Addition of ananthraquinone does not affect DNAmelting temperature as it would in case of intercalation. Besides, thequinone-DNA complex formation is affected by the environment, for instance, ionic strength of medium,to a much higher extent than it is expected in intercalating binding. [37] It is also supported by com-parison of fluorescence in anthraquinone complex with a double stranded DNA (dsDNA) and a singlestranded DNA (ssDNA), the latter of which expressed stronger quenching, unusual for intercalation. [36]An electron transfer in the CA-DNA complex can be initiated by light irradiation which activates trans-fer from G base to the chromophore or by structural rearrangement of the anthraquinone due to thehydrophobic environment of DNA. The binding constant (KB) of CA - DNA complex is 5 x 105 (M basepair)-1 (Figure 1.15 a), which is close to KB of well-studied complexes of DNA and doxorubicin or dauno-mycin (anthracyclines with structural similarities to CA): 1.5 x 105 (M base pair)−1 and 7 x 105 (M basepair)-1, respectively. Values around this range indicate strong interaction of a chromophore with DNA.CA complex formation with DNA can be detected optically: usually, the fluorescence emitted by CA ispredominant in the blue region (around 15,000 cm-1); however, in complex with DNA it shifts to the or-ange region (around 24,000 cm-1) (Figure 1.15 b). [35] As a complex between an anthraquinone andDNA forms, it leads to a shift in peak potential [24, 36] and either decreased peak current in general [36]or only in case of mismatched C-A and G-A base pairs [24]. It has been observed that the more doublestranded DNA is added to anthraquinone solution, the bigger is the loss of the electrochemical signal.What is more, due to a competitive binding, another compound can be used to displace anthraquinonein a complex with DNA, hence, changing the electrochemical response of the system. [38]131.2. Carminic acid(a) (b)Figure 1.15: (a) Bound CA concentration dependence on DNA concentration. (b) Emitted fluorescencespectra of CA and CA - DNA complex (15 000 cm-1 to 16 000 cm-1 = 667 nm to 625 nm, 23 000 cm-1 =435 nm). Reprinted from [35], Copyright (2015), with permission from Elsevier.1.2.4 CA as an electrochemical molecular beacon (ECMB)Although, the quinone in CA molecule undergoes a redox reaction, CA dimers have been found tobe electrochemically inactive. The ability to switch on and off the electrochemical signal is a propertysuitable for a MB. It has been, therefore, successfully used as a signaling molecule in electrochemicalmeasurements, in other words, as an electrochemical molecular beacon (ECMB). [9, 8]In one of the studies, CA ECMB was used in combination with a human porphobilinogen deaminasepromoter based sequence which also included self-complementary stem section sequences that werenot related to the target. The whole sequence can be covalently attached to CA on both of its ends and,as the hairpin loop forms, the attached CA dimerizes and is electrochemically inactive. Upon hybridiza-tion with a complementary strand, the hairpin loop opens, separating the CA monomers and switchingthe electrochemical activity back on. [8] Based on this mechanism (Figure 1.16) [8], an electrochemi-cal sensor for a quantitative detection of HIV-1 has been developed by using specific to it nucleic acidsequences [10].141.2. Carminic acidFigure 1.16: Mechanism of CA as ECMB in a hairpin loop form. Reprinted from [8], Copyright (2015),with permission from Elsevier.A similar study was done for detection of thrombin in which the nucleic acid sequence used wasbased on the anti-thrombin aptamer sequence. Apart from the stem section as described before, therewas a linker hybridized to the sequence which would fix it to a set position on a magnetic nanobead,forming a stem loop. The ECMB mechanism of this case (Figure 1.17) is almost identical to the onedescribed above where the loop of nucleic acid sequence keeps CA dimerized and the hybridization toa complementary strand results in separated CA monomers. [9]Figure 1.17: Mechanism of CA as ECMB in a stem loop form. Reprinted from [9], Copyright (2015), withpermission from Elsevier.The electrochemical activity of CA as ECMB, regardless of the attached nucleic acid sequence, isalike. Cyclic voltammetry (CV) of CA (Figure 1.18) in 0.1M pH 7.3 phosphate buffer and 1-1.5 M NaCl151.2. Carminic acidand optional 1 M MgCl solution on graphite and naphion/graphene electrodes shows a distinctive redoxpeak pair around -0.6 mV and an oxidation peak around +0.5 mV. The same peaks appear in CV ofCA-ECMB with a hybridized to it complementary nucleic acid sequence. However, dimerized CA in ahairpin loop or stem loop formation of CA-ECMB results in no redox activity. [9, 8, 10]Figure 1.18: CV of CA as an ECMB in its dimer form (CAs-MB), as ECMB in its monomer form (CAs-MB/thrombin complex) and in its free form (CA) on a graphite electrode in pH 7.3 phosphate buffersolution with 1 M NaCl and 1 M MgCl at scan rate of 100 mV/s. Reprinted from [9], Copyright (2015),with permission from Elsevier.Differential pulse voltammetry (DPV) (Figure 1.19) provides a comparison of the obtained electro-chemical signal for CA-ECMB with and without hybridization of various complementarity nucleic acidsequences. Although there is some current from the dimerized CA in CA-ECMB, it is only a small frac-tion of the signal of CA-ECMB hybridized to a complementary sequence. Addition of non-complementarynucleic acid sequences has no effect on the CA-ECMB signal, whereas, one base pair mismatch resultsin a 50% current loss in comparison to a complementary sequence. [8, 10, 9]161.3. Scope of the projectFigure 1.19: Differential pulse voltammetry (DPV) measurements of 1 µM CA as ECMB (4 - dash-dotpurple line) and the CA-ECMB hybridized to a complementary (1 - solid red line), a complementary witha one base mismatch (2 - dotted blue line) and a non-complementary (3 - dashed green line) nucleicacid sequences (of 5 µM concentration each) on a graphite electrode in pH 7.3 phosphate buffer solutionwith 1 M NaCl. DPV parameters were set to pulse width 0.5 s, pulse interval 0.2 s, amplitude 50 mVand potential step 50 mV. Reprinted from [8], Copyright (2015), with permission from ElsevierThe limitations for the use of CA as ECMB use include temperature and pH of the medium. [9]Although CA-ECMB can be used at elevated temperatures; upon reaching its melting temperature ofabout 52-53oC, the closed loop structure opens, regardless of hybridization CA becomes a monomerand the complex does not function as an ECMB anymore. [9, 8, 10] Besides, CA cannot be used asECMB at pH values below 6 due to protonated CA difficulties to form dimers but no issues with CAdimerization were found at neutral and basic pH values. [9]1.3 Scope of the projectSpecific objectives for this thesis are to investigate properties of CA and evaluate its suitability as aprospective ECMB for TB detection and differentiation between its strains. CA is to be electrochemi-cally characterized under various conditions, including, its concentration and electrolyte pH, as well as,analyzed by a number of electrochemical methods (CV, SWV, rotating disk electrode). The differencesin electrochemical activity between the four CA species are also to be investigated. For the prospectiveapplication of CA in a multiplex detection system along with other compounds as additional ECMBs, thesystem requires a significant difference in either redox potentials or the kinetics of the compounds used.Therefore, a method for analysis of CA kinetics is to be developed and applied experimentally. All dataon electrochemistry of CA is expected to form a reference guide for the future studies.171.3. Scope of the projectThe work on the thesis has to take into account the concern for the bigger picture which is to developa single chamber microfluidic device for multiplex detection of TB and differentiation of its strains. This isplanned to be achieved in collaboration between synthetic chemistry, electrochemistry and engineeringlaboratories. The synthetic chemistry lab is responsible for synthesis of a family compounds based onCA and for their coupling to DNA. Electrochemistry is then used to characterize these compounds andtheir complexes with ssDNA, in free form and hybridized to complementary and non-complementarystrands, in order to evaluate their suitability as ECMBs and for use in a multiplex detection system.Meanwhile, the engineering lab is developing a portable microfluidic device that would purify and am-plify nucleic acids and use an electrochemical system to detect TB and differentiate between its drug-susceptible and drug-resistant strains.18Chapter 2Electrochemistry Background2.1 Theory2.1.1 Electrode-solution interfaceElectrochemical reactions occur at the electrode-solution interface. It can be described to behave like acapacitor [39, 40] – as a current passes through two conducting plates separated by a dielectric material– charge accumulates on one plate with a counter-charge on the other plate until the following conditionsare fulfilled:C=qE(2.1)where C is capacitance (measured in farads, F), q is a charge that is accumulated on the capacitor(measured in coulombs, C) and E is potential difference between the capacitor plates (measured involts, V). [39]In the same way, there is a charge caused by lack or excess of electrons on the electrode (often ametal) surface, qM, balanced by the charge caused by excess of ions in solution near the surface ofthe electrode, qS. Although the magnitudes and signs of these values might vary and depend on thepotential difference between these two environments, at equilibrium qM is equal to -qS. [39] This modelof the electrode-solution interface (Figure 2.1) is generally called an electrical double layer. It consistsof the inner Helmholtz plane (IHP), located at distance x1 from the electrode, and mainly consists ofsolvent molecules and in some cases also of specifically adsorbed ions, and the outer Helmholtz plane(OHP) which is the closest distance to the electrode (x2) that the non-specifically adsorbed solvatedions can achieve. Distances of both planes are measured from the electrode surface to the center ofthe solvated ions. The diffuse layer extends from the OHP, where the excess charge is the highest, tothe bulk solution, where no excess charge exists. Its thickness depends on the concentration of ions insolution [39, 40] and is measured as 1.5 times the Debye-Hu¨ckel length:192.1. Theoryκ−1 =‚ϵrϵ0kBT2C∗z2 e2Œ1/2(2.2)where ϵr is relative dielectric permittivity of the electrolyte, ϵ0 - dielectric permittivity in the vacuum, kB- Boltzmann constant (1.38 × 10−23 m2.kg.s-2.K-1), T - temperature, C* - electrolyte bulk concentration,z - ionic charge and e - elementary charge. Apart from these parameters, the double layer thicknessalso depends on the potential. The larger is the difference between potential of zero charge (at whichthere is no charge on the electrode) and the electrode potential, the smaller is the thickness. [40]Figure 2.2 shows a change in potential over the double layer and it can be seen that the biggest drop inpotential occurs between the metal surface and OHP, followed by a gradual change in potential over thediffuse layer. [39] Therefore, the double layer must be taken into account when considering the electrodekinetics.Figure 2.1: Schematic representation of electrical double layer. Reprinted with permission from [39].Copyright (2015) John Wiley and Sons.202.1. TheoryFigure 2.2: Potential drop across the electrical double layer. Reprinted with permission from [39]. Copy-right (2015) John Wiley and Sons.2.1.2 Electrochemical processesFaradaic and nonfaradaic processesFaradaic and nonfaradaic are two types of processes that occur at the electrode surfaces. The formertype includes reactions with charge transfer between the electrode and the solution – the reduction andoxidation (redox) reactions. Faradaic reactions are usually the ones that carry the value in chemicalanalysis. Nonfaradaic processes, for example, adsorption, also result in current flow with a changein potential; however, no charge transfer occurs at the electrode-solution interface. Apart from faradaiccurrent, charging current due to capacitance also occurs at the electrode. Since all these current sourcescontribute to the total current, it is important to distinguish faradaic from nonfaradaic currents in order toevaluate the kinetics properly and to make analytical measurements of concentrations of redox species.[39]212.1. TheoryReversibilityA redox reaction is considered to be electrochemically reversible if the concentration of redox speciesat the surface of the electrode at a given potential follows Nernstian behavior, defined by the Nernstequation [39, 40]:E= E0′+RTnFnCO(= 0)CR(= 0)(2.3)where E0′is a formal potential which describes the equilibrium state of the redox reaction, R - thegas constant (8.314 J.mol-1.K-1), T - temperature (measured in K), n - number of electrons involved inelectron transfer of the reaction, F - the Faraday constant (96 485 s.A.mol-1), and CO(x = 0) and CR(x =0) are the concentrations of redox species at the surface of the electrode. [39]Generally, reversible reactions are characterized by fast kinetics as equilibrium is quickly establishedupon changes in conditions. In irreversible redox reactions oxidation or reduction occurs magnitudesfaster than the opposite process, resulting in slow kinetics. Anything between the fast and the very slowkinetics is considered to be quasi-reversible. [40]Voltammetry can be used to determine the reversibility of a redox couple by comparison of anodicand cathodic peak currents, ipa and ipc. Generally, if the ratio of both peak currents is close to 1:1 ofequal concentration redox species, redox reaction is considered to be reversible. The difference in peakpotentials for a reversible system is 2.3RTnF or 59/n mV at 25oC. [39]Electrochemical kineticsIn a redox reaction between two redox species O and RO+ nekƒŠkbR (2.4)kƒ and kb are the rate constants (measured in cm.s-1) for the reduction (forward) and oxidation (back)reactions. Knowing the concentration of O and R forms (CO and CR), they can be used to obtain the netreaction rate for O to be reduced to R with Equation 2.5. [39, 40]υnet = kƒCO(= 0)− kbCR(= 0) (2.5)222.1. TheoryE0’ describes a potential at equilibrium where CO and CR are at constant concentrations and kƒ andkb have the same values. To evaluate kinetics at this potential, the standard rate constant, k0, is used.It essentially describes the time required for a redox couple to establish equilibrium, and the larger k0is, the faster is the redox reaction. As rate constants are potential dependent, at potentials other thanE0’, k0 relationship to kf and kb is described by Equation 2.6 and Equation 2.7:kƒ = k0ep[−αnƒ (E− E0′)] (2.6)kb = k0ep[(1−α)nƒ (E− E0′)] (2.7)In the equations above, α is a transfer coefficient and ƒ = FRT . The transfer coefficient describes thesymmetry of the energy barrier of a redox reaction and 0 ≤ α ≤ 1 is the range of its value. [39, 40]As it can be seen in Figure 2.3, a larger α value for reduction processes means a steeper slope ofthe energy barrier and, hence, slower reaction, and vice versa. A value of 0.5 assumes a symmetricalenergy barrier for reduction and oxidation processes and, if the value of α is not known, it is a goodapproximation that can be used in calculations. [39]Figure 2.3: Energy barrier symmetry for different α values. Dashed line represents energy barrier atmore positive potential. Reprinted with permission from [39]. Copyright (2015) John Wiley and Sons.For a faradaic process, current, potential and k0 are related by the following equation:= nFAk0hCo(0, t)e−αƒ (E−E0′ )−CR(0, t)e(1−α)ƒ (E−E0′ )i(2.8)where, in addition to the previously known terms, A is the area of the electrode (measured in cm2).Exchange current, i0, is the value of a reduction or oxidation current flowing at equilibrium when the net232.1. Theorycurrent is 0. At these conditions, it is directly proportional to k0 [39]:0 = nFAk0C (2.9)This leads to the Butler-Volmer equation (Equation 2.10) which is correct at the conditions whensurface concentration is close to bulk concentration.= 0”e−αƒη− e(1−α)ƒη—(2.10)This equation introduces overpotential, η, which is a potential difference between applied and equi-librium potentials (η= E− Eeq). The larger the overpotential, the larger the magnitude of a reduction oran oxidation process occurring at the electrode. Tafel behavior describes the situation in which one ofthe redox processes is so large that the other term can be ignored. As a logarithm of current is plottedagainst overpotential, a Tafel plot is obtained (Figure 2.4). It has a slope of (1−α)nF2.3RT and−αnF2.3RT for oxida-tion and reduction processes, respectively, which can be used to determine the transfer coefficient. Dueto decay of Tafel behavior closer to the equilibrium potential, it loses its linearity at small overpotentials;however, extrapolation to equilibrium potential can be used to obtain a value of i0. However, it is difficultto observe Tafel behavior when current is limited by mass transfer. [39, 40]Figure 2.4: Example of a Tafel plot. Reprinted with permission from [39]. Copyright (2015) John Wileyand Sons. [39]Apart from the actual electron transfer, there are several processes that also affect the rate of a redox242.1. Theoryreaction – reactant transfer from bulk solution to the surface of the electrode where reaction can takeplace (mass transfer) and other reactions occurring either with the compound itself (e.g., dimerization)or on the electrode surface (e.g., adsorption). The overall reaction rate depends on the rates of all ofthese processes, the slowest of which is a rate determining step which affects the rest of the kinetics bymaking the reaction only as fast as the limiting step. In simpler reactions which consist of only electrontransfer and mass transfer, the latter is an important issue to consider. [39]Mass transportAs the charge transfer of a redox reaction occurs at the surface of an electrode, it is necessary to bringthe reactant from the bulk solution to the electrode surface first. It can be done through migration,diffusion and convection processes. Migration is caused by a difference in potentials. As the majorpotential drop occurs across the double layer, it has only a small impact on the mass transfer and canbe eliminated with the use of a high concentration of additional electrochemically inactive electrolyte.Diffusion is controlled by a difference in concentrations which is the main driver for mass transfer duethe lack of reactants at the surface of the electrode (CO = 0), while the concentration in the bulk solutionis high (C*O>> 0). If diffusion is limiting the current, redox charge transfer is occurring at the maximumrate that concentration of reactants at the surface of the electrode allow and all species react as soon asthey reach the surface. In order to study redox reaction kinetics, mass transfer should be increased toavoid limiting the redox current. This can be accomplished by convection which is caused by a physicalflow of solution, for example, solution stirring or use of an RDE. [39, 40]2.1.3 Electrochemical cellAn electrochemical cell consists of two or, usually, three electrodes. [39] The redox reaction of interestoccurs on a working electrode (WE). [39, 40] The potential applied to a WE is measured against anelectrode of a known, constant composition and, therefore, fixed potential called a reference electrode(RE). [39, 40] As more negative potentials are applied, electrons flow from the WE to the solution andreduction occurs (reduction current); on more positive potentials, electrons flow to the WE causing ox-idation (oxidation current). A counter electrode (CE) is often used as a third electrode and it forms acircuit with WE, resulting in a current passing between these WE and CE. [39]252.2. Methodology2.2 Methodology2.2.1 Linear potential sweepOne of the convenient ways to obtain a current-potential profile of a redox reaction is by using a linearpotential sweep voltammetry where the applied potential is changed in a linear fashion at a certain ratecalled the sweep rate, υ (measured in mV/s), and the current response is measured.As the double layer is charged, capacitive current is observed in the linear potential sweep. Figure2.5 shows electrochemical cell components as parts of a circuit where Cd is capacitance of the doublelayer, CSCE - a capacitance of a CE (RE/CE combination in two electrode configuration or only CE inthree electrode configuration) and RS - resistance of the solution. Due to the minor impact of CSCE to thetotal capacitance (as CE usually has a much larger surface area than the WE), it can be further ignored.Figure 2.5: Schematic representation of a cell as a circuit. CSCE represents a capacitance of a SCERE/CE combination in a two electrode configuration (pictured on the left) which in three electrode con-figuration is replaced by CE capacitance. Reprinted with permission from [39]. Copyright (2015) JohnWiley and Sons.Charging current, coming from the double layer capacitance, appears as the scan is started andincreases until it reaches a steady value of υCd. As potential sweep changes direction, the capacitivecurrent changes sign and increases again until it obtains a steady value of -υCd. The charging currentdependence on the applied potential is represented in Figure 2.6 and, as shown, it is directly proportionalto the sweep rate.262.2. MethodologyFigure 2.6: The potential applied in linear sweep (top left) and resulting charging current versus time(bottom left) and potential (bottom right). Reprinted with permission from [39]. Copyright (2015) JohnWiley and Sons.Faradaic current is usually the current of interest and it is caused by electron transfer at the surfaceof the electrode. As the applied potential reaches a negative enough value, reduction process producesa reduction current which increases as the potential measured against the RE becomes more negative.However, the concentration of a compound at the surface that can be reduced also decreases and, atone point it reaches zero as mass transfer is not capable of bringing it to the surface of the electrodefast enough. Any further generated reduction current is diffusion limited. Oxidation current is producedby oxidation process at more positive potential in a similar manner.CV is a method where the linear sweep of potential is cycled between two potentials resulting inconsecutive reduction and oxidation cycles. Figure 2.7 shows the potential applied and the resultingvoltammogram.272.2. MethodologyFigure 2.7: (a) Applied potential and (b) the resulting CV. Reprinted with permission from [39]. Copyright(2015) John Wiley and Sons.The obtained faradaic peak current in CV is strongly dependent on the sweep rate and it can bedescribed by the following equation:p =2.69× 105n3/2AC∗OD1/2Oυ1/2 (2.11)For irreversible systems, this equation is corrected to:p =2.99× 105α1/2AC∗OD1/2Oυ1/2 (2.12)In both cases, since the number of electrons, transfer coefficient, concentration in bulk solution, areaof the electrode and diffusion coefficient are not changing with time or potential, it is obvious that, forpure faradaic current, peak current is directly proportional to the square root of sweep rate.However, peak potential relationship with the sweep rate (Equation 2.13) is much more complicatedwhich does not allow any quick conclusions about the system. [39]Ep = E0′ −RTαF0.780+ n D1/2Ok0!+ nαFυRT1/2 (2.13)In addition, the relationship between the peak potential (Ep) and the potential where current is half ofthe peak current value (Ep/2) can be used to determine either number of electrons involved in a reactionin a reversible system (Equation 2.14) or the transfer coefficient of a redox reaction in an irreversiblesystem (Equation 2.15):282.2. Methodology|Ep− Ep/2|=2.20RTnF(2.14)|Ep− Ep/2|=1.857RTαF(2.15)2.2.2 Pulse voltammetryPulse voltammetry is another electrochemical technique that is based on measuring current obtainedfrom an applied potential that is modified in a stepwise manner. Any change in potential results in achange in current response; however, faradaic current and charging current behave differently. TheCottrell equation (Equation 2.16) predicts a decrease in faradaic current after charging in square rootrelation to time:(t) =nFAD1/2O CO∗pi1/2t1/2(2.16)where t is time (measured in s), and the initial current, on potential pulse application (t = 0), depends onredox reactions and mass transfer limitations. Whereas, charging current is caused by the capacitivenature of the double layer and decays exponentially over time. Since the decrease in faradaic currentover time is different than the decrease in capacitive current, pulse voltammetry enables minimal impactof the capacitive current if it is measured when sufficient time has passed since potential was applied.This method allows higher resolution measurements of faradaic reactions and some techniques basedon pulse voltammetry are described further in this section. [39]Staircase CVStaircase CV is basically very similar to a linear potential sweep CV. The applied potential is cycledbetween two potential values and it is changed in a stepwise manner by the amount ΔE every timefraction, τ, which can be as small as microseconds if required (Figure 2.8). The sweep rate can beobtained by ΔEτ . If potential steps are small enough (ΔE below 5 mV), the staircase CV can be treatedjust like linear sweep CV with all the applicable theory for faradaic currents. [39]292.2. MethodologyFigure 2.8: Potential change in staircase CV with each applied potential step noted as one pulse cycle.Reprinted with permission from [39]. Copyright (2015) John Wiley and Sons.Differential pulse voltammetryDPV is a combination of potential steps in one pulse cycle. For the most of the time, potential is heldat its initial base value, then it is increased by a certain value (known as modulation amplitude), heldfor a fraction of a time (modulation time), and returned to a new base potential which is different fromthe initial base potential (see, Figure 2.9). The resulting current is presented in Figure 2.10. The setmodulation amplitude, modulation time and total interval time stay constant throughout the experimentand the difference in both base potentials is defined as step potential.Figure 2.9: Potential steps in DPV. Reprinted from [41] with permission from Metrohm Autolab B.V.302.2. MethodologyFigure 2.10: Current response to the applied potential steps. Reprinted with permission from [39].Copyright (2015) John Wiley and Sons.Current is measured two times during each pulse cycle, just before the potential step modulation andjust before the drop to the base potential, and the difference in these two values, δi, is plotted againstthe potential, resulting in differential pulse voltammogram (Figure 2.11). In case of very negative orvery positive potentials, where only a reduction or oxidation process has to be taken into account, asmall change in potential causes a minimal change in current. However, at potential values closer to theequilibrium potential, a slight change in potential may considerably increase the opposite redox process,hence, resulting in a significantly larger difference in currents. [39]Figure 2.11: Example of DPV. Reprinted with permission from [39]. Copyright (2015) John Wiley andSons.312.2. MethodologySquare wave voltammetrySWV is a special case of DPV. The main difference is the length of time between any potential pulses- for SWV all time intervals are of the same length, making it possible to set a potential step frequency.In the first pulse cycle, SWV starts at Ei to which a half of modulation potential is added as a pulse,then potential is reversed by a full value of modulation potential so that it reaches a new base potential(Figure 2.12). Ei is the average potential between these two potential pulses. In further pulse cycles,SWV follows DPV potential pulse profile. The current is sampled two times per pulse cycle - just beforeeach potential step.Figure 2.12: Potential profile in SWV. Reprinted with permission from [39]. Copyright (2015) John Wileyand Sons.Just like in case of DPV, there is a small or almost no difference in current measured between forwardand reverse samples at high overpotentials, and significant difference near equilibrium potential (Figure2.13). As both sample values might carry a relevant analytical information, they are sometimes plottedalongside the δi. [39]322.2. MethodologyFigure 2.13: Example of SWV. Forward (ψf) and reverse (ψr) currents represented with dashed lines anddifference in these currents with a solid line (ψ is a normalized dimensionless parameter of the current).Reprinted with permission from [42]. Copyright (2015) American Chemical Society.Although CV is generally a more popular electrochemical technique, SWV has lower detection limitsdue to the method effectively removing background and it is more suitable for measurements of severalredox compounds in the same solution as the detection sensitivity is also increased. [39]2.2.3 Rotating disk electrodeAs it has been mentioned previously in this chapter, reaction rates are often mass transfer limited. Whendiffusion is not fast enough, convection can be added in order to bring necessary redox species to thesurface of the electrode. A rotating disk electrode (RDE) is used for this purpose. Generally, an RDE isa planar surface electrode which consists of a conductive material rod encased in an inert material. It isattached to a shaft that rotates the electrode at a set frequency, ƒ . Angular velocity, ω= 2piƒ , is moredescriptive, therefore, it is used in the analyses.Figure 2.14 schematically describes the flow of solution and its velocity in an RDE experiment. Asthe electrode rotates, it carries a solution layer with it and, by creating a flow in the solution, it brings thereactive species from the bulk.332.2. MethodologyFigure 2.14: (a) Mass transfer velocities and (b) flows at the RDE. Reprinted with permission from [39].Copyright (2015) John Wiley and Sons.Although it is then brought to the very surface of the electrode by the diffusion, the rotation decreasesthe diffusion layer thickness (Equation 2.17):δO = 1.61D13Oω− 12ν−16 (2.17)where δO is the diffusion layer thickness and ν represents kinematic viscosity (0.008844 cm2/s for0.1 M KCl in water). [39, 40]Instead of the current - potential relationship forming a peak shape as it is in the absence of rotation,the current measured with the electrode rotation reaches and maintains a steady-state value, il, (Figure2.15) which is explained by a constant diffusion of reactant to the electrode surface through δO. [39]342.2. MethodologyFigure 2.15: Example of current vs. potential measurement using RDE. [43] Adapted by permission ofThe Electrochemical Society.The Levich equation (Equation 2.18) is applied to a purely mass transport limited currents and itpredicts a proportional relationship between the limiting, steady-state current and the square root ofangular velocity. If the relationship between these two values is not linear and/or the plot does notintersect 0 on the x and y axes crossing, it is assumed that the diffusion limited current potential has notbeen achieved. = 0.620nFAD23ω12ν−16C (2.18)Since the total current measured includes the kinetic current, iK, it leads to a Koutecky´-Levich equa-tion [39, 40]:1=1K+1=1k+‚10.620nFAD23ν−16CŒω−12 (2.19)As it is plotted for several potentials, a value of the kinetic current can be obtained by extrapolating1/  to ω−12 = 0 (Figure 2.16). The more kinetics are limited by diffusion, the closer 1/ K value is to 0. Tofurther evaluate the kinetics at each potential, iK is related to the kf though the following equation [39]:K = FAkƒ (E)C∗O(2.20)352.2. MethodologyFigure 2.16: Example of a Koutecky´-Levich plot. E1 represents slower charge transfer than at E2.Reprinted with permission from [39]. Copyright (2015) John Wiley and Sons.A log of the found iK values can also be plotted against the η to obtain a Tafel plot (as in Figure 2.4).i0 value can be extracted from the Tafel relationship by extrapolating the slope to η = 0 which, in its turn,is directly related to k0 (see, Equation 2.9) and enables kinetic characterization of the redox reactivespecies.36Chapter 3Experimental3.1 MaterialsCA used in most of the experiments was from Dudley Corporation (assay 95%). CA stock solutions weremade at higher concentrations than required for electrochemical measurements (10 mM to 60 mM) andthen added to the electrolyte in the electrochemical cell to the desired concentration (1, 10, 100 μM, 1mM). The stock solutions were stored in a refrigerator.Deionized (DI) water produced from a Millipore Milli-Q Integral 5 water system (18.2 MΩ ·cm, totalorganic carbon of less than 4 ppb).Phosphate buffer solution was used as an electrolyte and it was made of monobasic sodium phos-phate (Fisher Scientific, Enzyme grade) and dibasic sodium phosphate (Fisher Scientific, ACS grade)combination to a total concentration of 0.01 M (see, Table 3.1). In order to decrease or increase the pH(from pH 1.1 to pH 10.5), respectively, nitric acid (Fisher Scientific, ACS grade, and BDH, ACS grade)and sodium hydroxide (Fluka, ACS grade) were added to the electrolyte solution. Sodium nitrate (Cale-don, assay 99.0%) was then added as a supporting electrolyte, amount of which was adjusted to obtaintotal ionic strength of 0.1 M (the ionic strength of the electrolyte at pH 1.1 was 3 times higher due tothe amount of acid added). Although electrolytes at pH 4.1, pH 5.4 and pH 10.5 have a low bufferingcapacity, addition of another buffer system at these pH values could introduce changes in electrochem-ical activity of CA, therefore, phosphate solution at these pH values was used with caution. Phosphatebuffer solution around pH 7 was also used in preparation of CA stock solutions.373.2. Electrochemical cellpH sodiumphosphate,monobasicsodiumphosphate,dibasicsodium nitrate nitric acid sodium hydroxide1.1 0.1204 g - - 2.164 mL 15.9 M -2.9 0.1200 g - 0.7392 g 3.000 mL 0.1 M -4.1* 0.1202 g - 0.7632 g 0.150 mL 0.1 M -5.4* 0.1156 g 0.0055 g 0.7590 g - -6.7 0.0545 g 0.0781 g 0.6698 g - -7.0 0.0475 g 0.0857 g 0.6619 g - -8.1 0.0051 g 0.1363 g 0.6136 g - -10.5* - 0.1417 g 0.5942 g - 1.7 mL 0.18 M* No or low buffer capacity expected.Table 3.1: Composition of electrolyte at the experimental pH values.Concentrated sulphuric acid (Sigma-Aldrich, ACS grade) in 1:1 ratio with concentrated nitric acid(Fisher Scientific, ACS grade) were used for glassware cleaning in a hot acid bath. Saturated calomelelectrode (SCE) was stored and used in the experiments immersed in a saturated potassium chloride(Fluka, ACS grade) solution. Cobalt hexammine chloride (Sigma-Aldrich, assay 99%) was used as ananalyte in the development and verification of kinetic analysis. Other materials used were argon gas(Praxair, >99.998%) and high purity butane gas (Power®, ultra refined).3.2 Electrochemical cell3.2.1 ElectrodesAll electrochemical experiments were performed in a cell with a three electrode configuration whereWE and CE were located in the same cell compartment filled with the electrolyte and connected to theRE through a salt bridge (see, Figure 3.1). For this purpose, glassy carbon (GC) RDE (Metrohm, diskdiameter 3mm, part number 6.1204.300) was used as a WE, platinum wire (0.5 mm in diameter) as aCE and SCE (Beckman, part number 511100) as a RE. Prior to use, the GC electrode was cleaned byrinsing in DI water and ethanol and, if a change in the baseline current was observed, it was additionallypolished on a polishing cloth with aluminum oxide powder (Metrohm, diameter 0.3 µm, part number6.2802.000) and then rinsed again. The platinum CE was flamed clean in a butane flame and rinsed inDI water.383.2. Electrochemical cellFigure 3.1: Electrochemical cell setup.3.2.2 Electrochemical cell setupGlassware used for sample preparation, storage and as a part of the electrochemical cell was washedin heated acid bath for at least 3 hours, then rinsed and filled with DI water, left to soak overnight andrinsed again with DI water prior use. The choice of the electrolyte pH values was based on fractionalconcentration of CA species at given pH (Figure 1.2) - either only one predominant species in solution(at pH 1.0, 4.1, 6.8, 10.5) or about 50:50 mixture of two species (at pH 2.9, 5.4, 8.1). A biologicallyrelevant pH 7.0 phosphate buffer solution was used for limit of detection (LOD) measurements. Beforeexperiments, the electrochemical cell was filled with phosphate buffer and oxygen was purged withargon bubbling for about 15 minutes. The salt bridge was then filled with phosphate buffer solution fromone side and saturated potassium chloride from the other side to make a connection with RE. Duringthe measurements, a steady argon atmosphere was maintained. All experiments were performed atroom temperature. After the blank measurements of base electrolyte were done, CA was added tothe electrolyte from the stock solutions to a resulting molarity of 1, 10, 100 µM and/or 1 mM. After anyaddition of CA, the solution was bubbled with argon for another 5 minutes.393.3. Instruments3.3 InstrumentsAll electrochemical measurements were carried out using an Autolab PGSTAT30 potentiostat. Potentio-stat control, data collection and basic analysis were performed through NOVA (version 1.10) software.Brinkman Metrohm 628-10 controller was used to set rotation rate of the RDE (Metrohm, part number1.628.0020). The measurements were performed in a Faraday cage. A Cole Parmer microcomputerpH-vision (model: 05669-20) pH meter was used to measure pH of the solutions prepared.3.4 MeasurementsCV was run at a number of sweep rates with sampling frequency set to 18.7 Hz for sweep rates of10 mV/s and 20 mV/s as it provided the least noise with an adequate number of sampled data pointsacross the potential. For 50 mV/s sweep rate, the sampling frequency was raised to 37.4 Hz and forhigher sweep rates to 93.5 Hz to ensure sufficient amount of data collected. Bandwidth was set to ahigh stability setting. For SWV measurements, step potential was set to 4 mV and amplitude to 25 mV.Staircase CV was the technique used for the RDE measurements and it was set to 5 mV step potentialand 1 s total interval between these steps, resulting in a 5 mV/s scan. The specific sets of CV sweeprates, SWV measurement frequencies and RDE rotations rates are listed in Table 3.2.Potential range was adjusted at each pH to compensate for the shift of the redox peaks of interest.Method Parameter Parameter valuesCV Sweep rate, mV/s 10, 20, 50, 100, 200, 500 and 1000SWV Frequency, Hz 2, 5, 10, 20, 50, 100 and 200RDE Rotation rate, revolutions per minute (rpm) 500, 1000, 1500, 2000, 2500 and 3000Table 3.2: List of the experimental values of CV sweep rates, SWV measurement frequencies and RDErotations rates.40Chapter 4Results and DiscussionThis chapter presents the influence of pH on the electrochemistry of CA, alongside with some unusualfindings on fully protonated CA in strongly acidic conditions. The electron transfer kinetics were deter-mined using an RDEmethod and CA was characterized, including its LOD, by the prospective diagnosticelectrochemical detection method - SWV.4.1 Carminic acid electrochemistry at different pHCA electrochemistry expresses a strong relation to pH. CVs of 100 µM CA (Figure 4.1) show the shiftof the redox peak potentials to more negative values with increasing the pH. The experimental potentialrange was adjusted accordingly with a special attention to CA at pH 1 due to an unusual behaviorwhich is described further in this chapter. A quasi-reversible nature of CA electrochemistry allows for anestimate of the formal redox potential, E0’, for each pH by determining the average potential betweenthe reduction and oxidation peak potentials, presented in Table 4.1.414.1. Carminic acid electrochemistry at different pHFigure 4.1: CV of 100 µM CA in different pH phosphate buffer solutions on GC WE (area 7.0 mm2),sweep rate of 20 mV/s. Potential range for CA in pH 1.0 phosphate buffer was kept smaller due tosignificant additional electrochemical activity observed at more positive potentials (see, Section 4.2).pH Ered (V) Eox (V) E0’ (V)1.1 -0.380 -0.342 -0.3612.9 -0.517 -0.473 -0.4954.1 -0.645 -0.600 -0.6235.4 -0.661 -0.617 -0.6396.6 -0.747 -0.690 -0.7198.1 -0.808 -0.738 -0.77310.5 - - -Table 4.1: CA redox peak potentials and the estimated formal electrode potential (E0’) in different pHenvironments obtained from CV in 100 µM CA using a GC WE, sweep rate – 20 mV/s.424.1. Carminic acid electrochemistry at different pHThe shift in the reduction peak potential was found to be on average 61 mV per pH unit (Figure 4.2).As a shift of 59 mV per pH unit describes 1 electron 1 proton or 2 electron 2 proton transfer reaction, itagrees with a quinone electrochemistry and suggests a 2 electron 2 proton redox reaction. This is truefor values around pH 7 which is the pH of interest for future application. Although the non linear responsearound pH 4 could be caused by lack of buffering capacity of the electrolyte at the given pH, the increasein slope to 93 mV per pH unit might also suggest several CA species (with different protonation levels)being involved in the reaction, thus, changing the electron proton ratio of the redox process.Figure 4.2: Peak potential and pH relationship of 100 µM CA in different pH phosphate buffer solutionson GC at a sweep rate of 20 mV/s.Over a range of pH values, CVs of various sweep rates were obtained. Figure 4.3 shows an exampleof a 100 µMCA at pH 6.6; current was normalized by dividing it by a square root of sweep rate due to thelinear relationship between these parameters as explained by Equations 2.11 and 2.12 (see, Section2.2.1). Similar data was gathered for CA at each pH studied, and the peak current was extracted fromthese plots.434.1. Carminic acid electrochemistry at different pHFigure 4.3: CV of 100 µM CA in pH 6.6 phosphate buffer at different sweep rates on GC (area 7.0 mm2).The data is normalized by dividing the current by a square root of sweep rate.The reduction peak current was plotted against the pH (Figure 4.4). A general trend of current de-creasing with higher pH values can be seen, with an exception pH values from 4 to 6 as it has beenobserved earlier. No redox reactions have been observed at pH 10.5 which corresponds to the electro-chemical activity of the fully unprotonated CA species.Figure 4.4: Peak current and pH relationship of 100 µM CA at different sweep rates on GC.The linear (or close to linear) relationship between peak current and the square root of sweep rate(Figure 4.5) indicates that the CA redox reaction is diffusion limited and no or little adsorption processes444.1. Carminic acid electrochemistry at different pHare occurring. Two distinctive slopes can be differentiated between pH values from 1.1 to 4.1 (with pH2.9 slightly differing due to electrode pretreatment effects) and pH values from 5.4 to 8.1. This suggeststhat there is a difference in electrochemical activity between the CA species (see, Figure 1.2) and itcorrelates with previous results for CA below and above pH 4.1.Figure 4.5: Relationship between current and square root of sweep rate for 100 µM CA at differentpH. Dashed lines represent two distinctive slopes fitted to the data. The current and the sweep raterelationship at the same conditions verifies the linearity with the square root.To investigate the electrochemical characteristics of CA with changing its concentration, series ofmeasurements at different concentrations of CA over a range of pH values were obtained. An exampleof this in pH 6.6 phosphate buffer is shown in Figure 4.6. This and similar measurements were used toextract data for every pH value studied.454.1. Carminic acid electrochemistry at different pHFigure 4.6: CV of different concentration CA in pH 6.6 phosphate buffer on GC (area 7.0 mm2) with asweep rate of 20 mV/s. CV for a 1 mM CA was scaled (1:5) to fit the rest of the data.The nearly linear relationship between CA concentration and peak current (Figure 4.7) enables apossible use of CA in quantitative analysis at concentrations below 1 mM. It also indicates the absenceof dimerization process at these concentrations. Due to the lower electrochemical activity at basic con-ditions, results for low CA concentrations were difficult to measure and, hence, they are not included.Figure 4.7: Relationship between peak current (log) and concentration of CA (log) at different pH on GCwith sweep rate of 20 mV/s.464.1. Carminic acid electrochemistry at different pHThe proposed mechanism of CA redox processes is similar to that of a general quinone (see, Section1.2.2, Figure 1.12) and is presented in Figure 4.8. In comparison to a quinone, CA fractions into severalspecies (“1” to “4” in Figure 4.8) depending on the pH; and each species can undergo an electrochemicalquinone reduction. Upon the reduction, the product (“5” to “8” in Figure 4.8) may further undergo an acid-base equilibria as the pKa values of the CA species not necessarily correspond to the pKa values ofthe reduced CA species. For instance, the shift in the peak potential of CA around pH 4 is 93 mV perpH unit, which indicates a 3 proton 2 electron redox reaction. In this case, according to Figure 4.8, acompound “2” is electrochemically reduced to a compound “6” in a 2 electron 2 proton reaction, andthen it is proposed to be protonated by proton to a compound “5”. However, since the mechanism is notclear, the compound “2” could also be reduced directly to the compound “5”. It is also not excluded thatthere might be other pathways that are not 2 proton 2 electron processes.Figure 4.8: Proposed CA redox mechanism. “1” to “4” are CA species in different pH media, “5” to “8”are the reduced forms of CA, and “R” represents the sugar group.474.2. Characterization of carminic acid electrochemical activity at low pH4.2 Characterization of carminic acid electrochemical activity atlow pHAn unusual electrochemical activity of CA can be observed at strongly acidic conditions. Apart fromthe usual quinone redox peaks appearing throughout the pH values, there are additional redox peaks atabout 200 mV more positive potential (Figure 4.9). This unusual electrochemical behavior is assumedto be somehow related to the fully protonated CA species which are present at every pH (from pH 1.0to pH 4.1) where the more positive peaks are observed.Figure 4.9: First cycles of 1 µM CA CV in a pH 1.0 solution on GC, sweep rate 10 mV/s.Figure 4.9 also shows that the more positive potential redox peaks only appear if CA is first reducedat -0.4 V. A more detailed investigation agrees with this finding. As CVs are run for two potential rangescovering the the more negative and the more positive peaks separately (Figure 4.10), there are no redoxpeaks at the positive potential in absence of the redox peaks at the negative potential regardless of CAconcentration. Besides, the appearance of the positive potential redox peaks was found to decreasethe peak current of the negative redox peaks. It might suggest an adsorption process of a CA reductionproduct to the electrode surface where these adsorbed species prevent CA from participating in its’quinone’ based redox reaction.484.2. Characterization of carminic acid electrochemical activity at low pHFigure 4.10: 1 µM and 100 µM CA in pH 1.0 phosphate buffer on GC at different potential ranges,covering only the more negative redox peaks of interest (on the left) or only the more positive redoxpeaks (on the right) potential range. (sweep rate 10 mV/s).Comparison of both redox processes at various CA concentrations (Figure 4.11) shows how theCA reduction peak current at the negative potential increases with the concentration but, at the sametime, the reduction peak current of the redox peak at the positive potential remains constant. Thissupports an idea that the adsorption process is involved at the positive potentials because the fullycovered electrode with a limited surface area cannot adsorb more redox active species regardless oftheir increased concentration. The reduction current of the more positive potential peak has been foundnot to be linear with the square root of sweep rate, instead, it has a close to linear relationship withsweep rate (Figure 4.12) which indicates that adsorption is quite likely.494.3. Study of the electron transfer kinetics using an RDEFigure 4.11: The change in ratio of the CA redox peaks in CV at pH 1.0 on GC, sweep rate 20 mV/s.Figure 4.12: Relationship between the current and the square root of sweep rate (on the left) and thesweep rate (on the right) for the redox peak at more positive potential of 1 µM CA at pH 1 on GC WE.4.3 Study of the electron transfer kinetics using an RDEThe kinetics of the CA redox process was studied for four CA species: H4CA at pH 1.1, H3CA− at pH 4.1,H2CA2- at pH 6.6 and HCA3- at pH 10.5. Electrochemical measurements using an RDE were obtainedfor all four CA species (Figure 4.13). As expected, no electrochemical activity of fully deprotonated CAwas observed at pH 10.5, hence, no further analysis is possible at this pH. At pH 1.1, pH 4.1 and pH 6.6the CA reduction current as a function of potential was extracted at each rotation rate for analysis usingthe Ketucky´-Levich equation (Chapter 2, Equation 2.19) as was done for cobalt hexammine (Appendix504.3. Study of the electron transfer kinetics using an RDEA).Figure 4.13: RDE reduction curves of 100 µM CA at pH 1.1 (top left), pH 4.1 (top right), pH 6.7 (bottomleft) and pH 10.5 (bottom right), all in phosphate buffer solution on a GC (area 7.0 mm2).The reduction current obtained at specific potentials is plotted as a Koutecky´-Levich plot (Figure4.14). The value of ik is determined by extrapolation of the 1/i versus 1/ω-1/2 to the y-axis for eachpotential studied. Tafel plots of log ik versus potential are shown in Figure 4.15. The slopes of the linearregion of these plots are used to extract the value of α since the Tafel slope for a reduction process is−αnF2.303RT . These results are summarized in Table 4.2. Redox of CA at pH 1.1 has no significant linearregion at low overpotentials and does not completely follow the Tafel behavior. Hence, the best linearregion of the plot was estimated and the Tafel slope and any kinetic parameters were derived from it.514.3. Study of the electron transfer kinetics using an RDEFigure 4.14: Koutecky´-Levich plot for 100 µM CA at pH 1.1 (top), pH 4.1 (middle) and pH 6.7 (bottom).524.3. Study of the electron transfer kinetics using an RDEFigure 4.15: Tafel plots for CA at pH 1.1 (top left), pH 4.1 (top right) and pH 6.7 (bottom) with fitted Tafelslopes (dotted lines).The values of E0’ were estimated from the CV at each measured pH (see, Table 4.1). and are usedhere to determine the i0 and ultimately the value of the standard rate constant, k0. The slopes of aKoutecky´-Levich plot are also used to estimate the value of the diffusion coefficient of the CA species insolution.The summarized results for CA at four pH values are presented in Table 4.2. The magnitude ofTafel slope decreases with pH which also lowers the value of α. In this way, the energy barrier is lesssymmetric and in favor of reduction reaction at higher pH values. However, α value close to 0.5 at pH1.1 means similar slopes of the energy barrier for oxidation and reduction processes. As regards thestandard rate constant, k0 decreases with an increase in pH for CA at pH 4.1 and pH 6.6. Throughout thischapter, CA has been shown to be less electrochemically active at higher pH values which is presentedin the trend of decreasing k0. For CA at pH 1.1, the standard rate constant is at least two times smaller53Square wave voltammetrythan expected for this trend; however, it is not clear whether this value is the result of the electrochemicalactivity of the fully protonated CA or the relative error. The logarithmic values of k0 are presented to verifythat the propagation of errors has more impact on the relative error than errors from the measurements.The measured diffusion coefficient for CA agree with the diffusion coefficients for the structurally similar2-AQMS and 2,6-AQDS (as stated in [27]). The measured values at pH 4.1 and 6.6 are relatively closeto each other; whereas, the diffusion coefficient of CA at pH 1.1 differs from the trend and is two timeslower than for the other pH values. Apart from the relative error, which is much smaller for the diffusioncoefficient, the unexpectedly low values for CA at pH 1.1 could be explained as an effect of the fullprotonation of CA; however, it needs further investigation for proper explanation.pH Tafel slope α log k0 k0 (cm.s-1) DO (cm2.s-1)1.1* -15.984 ± 1.398 0.466 ± 0.041 -2.83 ± 0.50 1.47×10-3± 168% 1.81×10-6 ± 18%4.1 -13.231 ± 0.269 0.386 ± 0.008 -2.38 ± 0.17 4.18×10-3± 50% 3.36×10-6 ± 18%6.6 -11.778 ± 0.146 0.344 ± 0.004 -2.75 ± 0.10 1.76×10-3± 32% 3.26×10-6 ± 18%10.5** - - - - -* Tafel-behaviour related parameters are approximated.** No electrochemical activity.Table 4.2: Summarized results of 100 µM CA kinetic parameters on a GC electrode.Square wave voltammetrySWV is the proposed method for detection of CA in the TB sensor being developed because of theincreased sensitivity towards redox reactions, in comparison to linear sweep methods. The redox of CAwas characterized by SWV to produce a reference for future studies. The effect of different frequencieson CA redox was further investigated in order to optimize the detection method. Since SWV is usuallypresented as a difference in current of two measurements against the potential applied, the differencein forward and reverse currents is further referred to as δi.Figure 4.16 shows a comparison of the four CA species at the lowest (2 Hz) and highest (200 Hz)frequencies measured. CA analyzed at different SWV frequencies shows a general trend of increasingcurrent with higher frequencies. Although this increase applies to all species at the various pH values,this change differs significantly between the species. CA measured using a 2 Hz signal appears asa relatively sharp peak of a similar size at any pH, except, pH 10.5 where no redox reaction was ob-served, agreeing with the previously described results. As the frequency is increased, the overall δialso increases for all pH values. This is expected because faradaic and nonfaradaic currents decay54Square wave voltammetryover time (where the background current decays faster) and higher measurement frequency results ina sampling done in a shorter period of time after the charging of the electrode. In addition, there arechanges observed in peak shapes and the ratio of CA peak currents at the different pH. While at 200 Hzthe prominent redox peaks are broader and the side reactions at about 100 mV more positive potentialbecome more pronounced, the SWV peak heights at pH 4.1 and pH 6.6 become smaller in comparisonto pH 1.0, and CA redox measured at pH 10.5 shows a small ’bump’ that cannot be seen at 2 Hz.Figure 4.16: SWV of 100 µM CA at different pH (a) at 2 Hz and (b) at 200 Hz frequencies (baselineadjusted between the pH values) on a GC (area 7.0 mm2).Higher SWV frequencies carry out more frequent sampling, and therefore, more forward and reversepotential steps per one second. Since fast electron transfer reactions can follow these modulationsat high frequencies while the slower reactions might not, SWV may be used to differentiate betweencompounds based on their electron transfer kinetics. Smaller and broader peaks at pH 4.1 and pH 6.6in Figure 4.16 suggest that their kinetics are slower than for CA at pH 1.0 which, to some extent, supportsthe previous findings where k0 was determined to be the lowest for CA at pH 6.6. One of the possibleexplanations for the unusual behavior of CA at pH 10.5 is the increase in δi at higher frequencies whichcould have made a very small, unnoticeable redox reaction signal to be more evident. In addition tothis, a larger period of time passes since the charging of the electrode until a signal measurement isdone at lower frequencies. This might be long enough for current to decay significantly, leaving theobtained δi close to a constant value throughout the potential range. Still, even at frequencies wherean electrochemical reaction can be observed, CA at pH 10.5 appears to have very slow kinetics incomparison to CA at other pH values.Figure 4.17 demonstrates a general trend of δi increasing with the SWV frequencies. Although55Square wave voltammetrysome electrochemical activity of CA at pH 10.5 can be observed at high frequency measurements, thisrelationship is clearly not linear and differs significantly from CA electrochemical behavior at any otherpH. The smallest difference between the peak δi in various pH solutions is at 2 Hz and the largest is at200 Hz. This suggests higher frequencies for differentiation between compounds with variance in theirkinetics but only if redox peak breadth is not limiting.Figure 4.17: Relationship between δi (log) and frequency (log) of 100 µM CA at different pH (SWV).As the same relationship is plotted for each of CA species separately (Figure 4.18), the concentrationeffect on δi can be also examined. The 100 μM and 1 mM fully unprotonated CA at pH 10.5 follows thesame tendency as in the previous findings for SWV and, hence, will not be further discussed. Theother CA species follow a similar trend as previously described - the δi increases at higher frequencies;however, the obtained slopes of the current-frequency relationship vary significantly not only amongthe CA species but also between concentrations of a single species. Thus, it can be observed thatSWV could be used for quantitative analysis at low frequencies where the difference in δi measuredis the largest, whereas, the current dependence on CA concentration decays at higher frequencies.This especially can be observed at pH 1.1 and pH 4.1. At pH 6.6, slopes are not as steep and at 1mM concentration the measured δi is almost independent of frequency. Unusual data is obtained for 1mM CA at pH 1.1 and for 1 µM CA at pH 6.6 where a decrease in δi is observed at 5 Hz and 200 Hz,respectively. The reason for this behavior is not certain and would require further studies.56Square wave voltammetryFigure 4.18: Relationship between δi (log) and frequency (log) of several CA concentrations at (a) pH1.1, (b) pH 4.1, (c) pH 6.6 and (d) pH 10.5.In spite of smaller δi values at low frequency SWV measurements and lack of CA redox peaks atpH 10.5 that can be detected at higher frequencies, the advantage of using low frequency SWV mea-surements involves much sharper redox peaks that allow simultaneous detection of several compoundswith the same kinetic properties and similar peak potentials. Meanwhile, the higher frequency measure-ments enable better differentiation between compounds with differing kinetics. Therefore, it is necessaryto tailor the measurement to the compounds used.57Square wave voltammetry4.3.1 Limit of detection (LOD)Limit of detection (LOD) represents the lowest concentration of an analyte that can be differentiated fromthe blank with a confidence of at least three standard deviations. [44] The LOD for CA was determinedin a biologically relevant pH 7.0 phosphate buffer solution to determine the lowest concentrations of CAthat could be measured in future studies. The obtained SWV for the 1 nM to 1 µM concentrations of CAat 2 and 200 Hz are presented in Figure 4.19. The results agree with findings in the previous section:the low frequency (2 Hz) SWV measurements result in a sharper peak (signal to noise ratio about 10:1)and the high frequency (200 Hz) SWV measurements result in an increase in peak δi and much broaderpeaks that lack definition (signal to noise ratio about 3:1). Besides, at 200 Hz SWV, lower concentrationsof CA (10 nM and 100 nM) have redox peaks that do not show at 2 Hz, possibly due to difference in timethat passes from the moment of electrode being charged and the sampling allowing the already smallcurrent decay more at the low frequencies. Therefore, the estimate of CA LOD lies between 10 nM and1 µM depending on the measurement frequency.Figure 4.19: LOD for CA at pH 7.0 at 2 Hz (on the left) and 200 Hz (on the right).58Chapter 5Summary and Future WorkCarminic acid, a prospective ECMB, was electrochemically characterized over a range of pH values. Itwas found to undergo an electrochemically quasi-reversible 2 electron 2 proton redox reaction with ashift of 61 mV per pH unit. Whereas, the redox potential of CA shifts about 93 mV per pH unit aroundpH 4 and might involve a 2 electron 3 proton reaction. Typically, CA redox reactions are diffusion limitedwith no significant adsorption processes present. The exception is the fully protonated CA (H4CA) whichshowed additional redox peaks around 0.2 Vmore positive from the general quinone redox peaks. Thesepeaks at the more positive potential are proposed to be caused by the adsorption of species createdafter initially reducing CA. The close to linear calibration curve at concentrations below 1 mM enables apossibility of further use of CA in quantitative analysis.The CA species can be differentiated based on their electrochemical activity. First of all, a peakpotential shift is observed with a change in pH which correlates with the separation of the CA species.The reduction peak currents of H4CA and H3CA- are relatively similar, while the peak current of H2CA2-is significantly lower. No redox current was observed of HCA3- at any measurements, except at the highfrequencies of SWV where it was still notably smaller than of any other CA species. Besides, an RDEmethod to evaluate the electrochemical kinetics has been successfully applied in practice. The electrontransfer kinetics of CA were determined at different pH values. Generally, the more deprotonated CA is,the slower the electron transfer kinetics. As CA at pH 1 does not completely follow Tafel behavior, it hasto be taken into account that the obtained kinetic data for this sample is an estimate. Regardless of theuncertainties, the standard rate constant and the diffusion coefficient are both at least two times smallerthan expected. This leads to a conclusion that the full protonation of CA causes unidentified processesthat have yet to be investigated.SWV is the proposed method of detection in the prospective device. It was found that the peakcurrent increases with CA concentration; however, for H4CA and H3CA- species, the current value athigh frequency sampling is as large for 1 µM CA as it is for 1 mM CA. Measurement frequencies alsohave an influence on the obtained current signal - lower frequency measurements result in sharp redox59Chapter 5. Summary and Future Workpeaks, whereas, higher frequency measurements enable a comparison of electron transfer kinetics ofdifferent compounds. The LOD for CA was found to be 1 µM at 2 Hz measurement and 10 nM at 200Hz measurement. The data obtained in this thesis can be used for reference in further studies.Future work includes a study of electrochemical activity, including kinetics, of other syntheticallymodified CA compounds and complexes as prospective ECMBs. This would allow evaluation of thesecompounds as ECMBs and their suitability for a use in a simultaneous detection system if a signifi-cant difference in redox potentials or kinetics is observed. Comparison between the ECMB specificallycoupled to a nucleic acid sequence and ECMBmixed at random with the same nucleic acid in an electro-chemical cell would help investigating the effect of the possible nucleic acid intercalation into an ECMBon electrochemically active-inactive switching properties of the ECMB. SWV frequencies and potentialsof ECMBs should be evaluated and appropriate values should be selected to be used in a multiplexdetection.Validation of each compounds’ suitability to ECMBmethod for nucleic acid detection is the necessarynext step. To achieve this goal, the electrochemical activity of ECMB would be examined on its own andin different combinations with DNA. The suggested specific combinations are ECMB coupled to a nucleicacid on both of its ends or, as a control sample, only on one terminus. Additionally, the electrochemicalactivity would be determined for these complexes exposed to a complementary, complementary with 1or 2 mismatched base pairs and non-complementary nucleic acid strands.A new method for electrochemical kinetic analysis can be implemented. Techniques of interest couldbe semi-integration of CV and kinetic data extraction fromSWV. The newmethod would facilitate analysisof electron transfer kinetics in systems where use of an RDE is restricted. Besides, the results obtainedfrom the other method could verify the accuracy of the currently applied method.A long term goal to achieve is characterization of the selected compounds and the detection pa-rameters with real TB samples, thus, validating the device suitable for TB detection and differentiationbetween its strains.60References[1] World Health Organization. http://www.who.int, World Health Organization, Accessed June2014.[2] Lawn, S. D. et al. The Lancet Infectious Diseases 2013, 13, 349 – 361.[3] Public Health Agency of Canada. http://www.phac-aspc.gc.ca/id-mi/index-eng.php,Accessed March, 2015.[4] Health Canada. http://www.hc-sc.gc.ca/fniah-spnia/diseases-maladies/tuberculos/index-eng.php, Accessed March, 2015.[5] Centers for Disease Control and Prevention. http://www.cdc.gov/tb/publications/factsheets/testing.htm, Accessed December 2014.[6] Narreddy, S.; Muthukuru, S. Apollo Medicine 2014, 11, 88–92.[7] Bechtold, T.; Fitz-Binder, C.; Turcanu, A. Dyes Pigm. 2010, 87, 194–203.[8] Wu, J.; Huang, C.; Cheng, G.; Zhang, F.; He, P.; Fang, Y. Electrochem. Commun 2009, 11, 177 –180.[9] Cheng, G.; Shen, B.; Zhang, F.; Wu, J.; Xu, Y.; He, P.; Fang, Y. Biosens. Bioelectron. 2010, 25,2265–2269.[10] Li, B.; Zhengliang, L.; Situ, B.; Dai, Z.; Liu, Q.; Wang, Q.; Gu, D.; Zheng, L. Biosens. Bioelectron.2014, 52, 330–336.[11] Reyes-Salas, O.; Juarez-Espino, M.; Manzanilla-Cano, J.; Barcelo-Quintal, M.; Reyes-Salas, A.;Rendon-Osorio, R. J. Mex. Chem. Soc. 2011, 55, 89–93.[12] Dapson, R. Biotech Histochem. 2007, 82, 173–187.[13] Kunkely, H.; Vogler, A. Inorg. Chem. Commun. 2011, 14, 1153 – 1155.61References[14] Dapson, R. Biotech Histochem. 2005, 80, 201 – 205.[15] Baranyovits, F. Endeavour 1978, 2, 85 – 92.[16] Lloyd, A. Food Chem. 1980, 5, 91 – 107.[17] Rasimas, J.; Blanchard, G. J. Phys. Chem. 1995, 99, 11333 – 11338.[18] Guin, P.; Das, S.; Mandal, P. Int. J. Electrochem. 2011, 2011, Article ID 816202.[19] Rasimas, J.; Blanchard, G. J. Phys. Chem. 1994, 98, 12949 – 12957.[20] Rasimas, J.; Berglund, K.; Blanchard, G. J. Phys. Chem. 1996, 100, 7220 – 7229.[21] Jorgensen, K.; Skibsted, L. J. Food Chem. 1991, 40, 25 – 34.[22] Armandariz-Vidales, G.; Martinez-Gonzales, E.; Cuevas-Fernandez, H.; Fernandez-Campos, D.;Burgos-Castillo, R.; Frontana, C. Electrochimica Acta 2013, 110, 628–633.[23] Domenech-Carbo, A.; Domenech-Carbo, M. T.; Sauri-Peris, M. C.; Gimeno-Adelantado, J. V.;Bosch-Reig, F. Anal. Bioanal. Chem. 2003, 375, 1169–1175.[24] Batchelor-McAuley, C.; Li, Q.; Dapin, S. M.; Compton, R. G. J. Phys. Chem. 2010, 114, 4094–4100.[25] Laviron, E. J. Electroanal. Chem. 1983, 146, 15–36.[26] Bailey, S.; Ritchie, I. Electrochim. Acta 1985, 30, 3–12.[27] Wipf, D. O.; Wehmeyer, K. R.; Wightman, R. M. J. Org. Chem. 1986, 51, 4760–4764.[28] Gomez, M.; Gonzalez, I.; Gonzalez, F. J.; Vargas, R.; Garza, J. Electrochem. Commun. 2003, 5,12–15.[29] Uchimiya, M.; Stone, A. Chemosphere 2009, 77, 451 – 458.[30] Rich, P. Faraday Discuss. Chem. Soc. 1982, 74, 349 – 364.[31] He, P.; Crooks, R. M.; Faulkner, L. R. J. Phys. Chem. 1990, 94, 1135–1141.[32] Alligrant, T.; Hackett, J.; Alvarez, J. Electrochimica Acta 2010, 55, 6507–6516.[33] Rich, P. Biochim. Biophys. Acta 2004, 1658, 165 – 171.[34] Trammell, S.; Lebedev, N. J. Electroanal. Chem. 2009, 632, 127 – 132.62[35] Comanici, R.; Gabel, B.; Gustavsson, T.; Markovitsi, D.; Cornaggia, C.; Pommeret, S.; Rusu, C.;Kryschi, C. Chem. Phys. 2006, 325, 509 – 518.[36] Gholivand, M. B.; Kashanian, S.; Peyman, J. Spectrochim. Acta, Part A 2012, 87, 232–240.[37] Qiao, C.; Bi, S.; Sun, Y.; Song, D.; Zhang, H.; Zhou, W. Spectrochim. Acta, Part A 2008, 70, 136–143.[38] Pandey, P.; Weetall, H. Anal.Chem. 1994, 66, 1236 – 1241.[39] Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.;John Wiley & Sons, Inc., 2001.[40] Scholz, F., Ed. Electroanalytical Methods: Guide to Experiments and Applications, 2nd ed.;Springer, 2010.[41] NOVA, Voltammetric Analysis Tutorial. Metrohm Autolab B. V.[42] Kounaves, S. P.; O’Dea, J. J.; Chandresekhar, P.; Osteryoung, J. Anal. Chem. 1986, 58, 3199 –3202.[43] Zurilla, R. W.; Sen, R. K.; Yeager, E. J. Electrochem. Soc. 1978, 125, 1103 – 1109.[44] Skoog, D. A.; Leary, J. J. Principles of Instrumental Analysis, 4th ed.; Sounders College Publishing,1992.[45] Hromadova, M.; Fawcett, W. R. J. Phys. Chem. 2000, 104, 4356–4363.[46] Ji, X.; Chevallier, F. G.; Clegg, A. D.; Buzzeo, M. C.; Compton, R. G. J. Electroanal. Chem. 2005,581, 249–257.63Appendix AKinetic Analysis of Cobalt HexammineCobalt hexammine was used for verification of the kinetic analysis due to its electrochemical similaritiesto CA (almost irreversible redox reaction and slow enough, comparable to CA kinetics) and the availabledata in literature [45, 46]. For this purpose, the usual set of electrochemical measurements (see, Chapter3) was run with 100 µM cobalt hexammine in pH 7 phosphate buffer solution, the data was then analyzedas shown below and compared to the literature values.Some parameters, like, Do and α, can be extracted from a CV (Figure A.1) that is run at differentsweep rates, using Equation A.1 and Equation A.2:|Ep− Ep/2|=1.857RTαF(A.1)p =2.99× 105α1/2AC∗OD1/2Oυ1/2 (A.2)The necessary for analysis data is presented in Table A.1.64Appendix A. Kinetic Analysis of Cobalt HexammineFigure A.1: CV of cobalt hexammine at different sweep rates.Sweep rate (mV/s) ip (A) Ep (V) Ep/2 (V)10 -4.55 ×10-7 -0.2862 -0.229220 -4.92 ×10-7 -0.2968 -0.239150 -5.18 ×10-7 -0.3122 -0.2540100 -5.23 ×10-7 -0.3253 -0.2667200 -5.61 ×10-7 -0.3409 -0.2838500 -7.17 ×10-7 -0.3741 -0.31531000 -1.05 ×10-6 -0.4060 -0.3462Table A.1: Data obtained from cobalt hexammine CVObtaining α value from Equation A.1 is relatively simple and it can be done for a single sweep rate.Solving the Equation A.2, on the other hand, requires at least a few sweep rates and a value of eitherα or Do to calculate the value of the other parameter from the slope of Equation A.2 (Figure A.2). Toevaluate these methods, values were taken either from Equation A.1 (α) or from RDE measurements(α or DO). The results obtained are summed up in Table A.2 at the end of the chapter.65Appendix A. Kinetic Analysis of Cobalt HexammineFigure A.2: Relationship of peak current and square root of sweep rate of cobalt hexammine.As regards RDE analysis, staircase voltammetry was used to measure current response to potentialat different rotation rates (from 500 to 3000 rpm) of the electrode in the same system as described above.The obtained reduction curves are presented in Figure A.3. Koutecky´-Levich equation [39] (EquationA.3) was then applied for kinetic analysis on RDE:1=1k+‚10.620nFAD23ν−16CŒω−12 (A.3)Figure A.3: Cobalt hexammine RDE reduction curves at 500 to 3000 rpm.66Appendix A. Kinetic Analysis of Cobalt HexammineFor convenience of calculation, the rotation frequency was transformed into angular velocity usingthe following equation:ω= 2piƒ (A.4)Figure A.4 shows a graph obtained by plotting Equation A.3 at different potentials; it shows that themore negative is potential, the higher is the kinetic current and the lower is the limiting mass transportcurrent. Kinetic parameters, like, kinetic current, ik , and diffusion coefficient, DO, are extracted from, re-spectively, the intercept and the slope. The reduction rate constant, kf, at each potential is then obtainedby using the following equation [39]:kƒ =kFAC∗O(A.5)Figure A.4: Koutecky´-Levich plot for cobalt hexammine.When log of kinetic current is plotted against the potential, a Tafel plot (Figure A.5) is obtained, and,as Tafel slope equals to −αF2.303RT for a reduction current [39], this also enables calculation of α value.67Appendix A. Kinetic Analysis of Cobalt HexammineFigure A.5: Tafel plot for cobalt hexammine.The area of linear relationship between kinetic current and potential in Figure A.5 suggests the po-tential range of interest (-0.38 V to -0.10 V) for which the reductuion rate constant, kf , and diffusioncoefficient, DO, were obtained and used for kinetic characterization of cobalt hexammine.The summarized kinetic parameters are presented in Table A.2.kf(cm/s) α DO(cm2/s)EquationA.1N/A 0.81 ±0.03 N/AEquationA.2N/A 0.51 ** 2.82 ×10-6*** and 4.75 ×10-6**RDE 1.10 ×10-4to 2.03 ×10-2 0.48 1.70 ×10-7to 8.79 ×10-6Reference* 9.5 ×10-6, 3.7 ×10-4, 2.1×10-3 and 1.3 ×10-2[45]0.51 [46] 5.6 ×10-6[46], 7.76 ×10-6, 8.16×10-6, and 8.57 ×10-6[45]* Data obtained from several gold substrates at different cobalt hexammine and electrolyte concentra-tions.** Data from RDE measurements substituted into Equation A.2.*** Data from Equation A.1 substituted into Equation. A.2.Table A.2: Summarized cobalt hexammine kinetic parameters obtained by different methods.In conclusion, CV at different sweeprates can be used to obtain DO and α values which are accuratewhen one of the values from RDE measurements is used in Equation A.2; however, calculation of kfis not straightforward and bare CV is not sufficient for kinetic analysis. Whereas, the results obtainedthrough the RDE measurements are very similar to the values found in the literature - kf and DO fit thereference data and α is just barely below it; any discrepancies can be explained by use of different work-ing electrode and cobalt hexammine and electrolyte concentrations. Therefore, RDEmeasurements areassumed to be valid for kinetic analysis of CA.68

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167174/manifest

Comment

Related Items