UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electrodeposited DNA monolayers on gold : creation, evaluation and optimization Leung, Kaylyn Kyra 2019

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2019_september_leung_kaylyn.pdf [ 27.17MB ]
JSON: 24-1.0379715.json
JSON-LD: 24-1.0379715-ld.json
RDF/XML (Pretty): 24-1.0379715-rdf.xml
RDF/JSON: 24-1.0379715-rdf.json
Turtle: 24-1.0379715-turtle.txt
N-Triples: 24-1.0379715-rdf-ntriples.txt
Original Record: 24-1.0379715-source.json
Full Text

Full Text

Electrodeposited DNA Monolayers onGold: Creation, Evaluation andOptimizationbyKaylyn Kyra LeungB.Sc., The University of British Columbia, 2013A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Chemistry)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)June 2019© Kaylyn Kyra Leung 2019The following individuals certify that they have read, and recommend to the Faculty of Graduateand Postdoctoral Studies for acceptance, the dissertation entitled:Electrodeposited DNA Monolayers on Gold: Creation, Evaluation and Optimizationsubmitted by Kaylyn Kyra Leung, in partial fulfillment of the requirements for the degree ofDoctor of Philosophy in Chemistry.Examining Committee Members:Dan Bizzotto (Chemistry), Co-SupervisorHua-Zhong Yu (Chemistry, Simon Fraser University), Co-SupervisorW. Russell Algar (Chemistry), Supervisory Committee MemberKaren C. Cheung (Electrical and Computer Engineering), University ExaminerDavid Chen (Chemistry), University ExaminerAlexis Valée-Bélisle (Chemistry, Université de Montréal), External ExaminerAdditional Supervisory Committee Members:Roman Krems (Chemistry), Supervisory Committee MemberHongbin Li (Chemistry), Supervisory Committee MemberiiAbstractDeoxyribonucleic acid (DNA) self-assembled monolayers (SAMs) consist of bioelectronic inter-faces that have been proposed for use as DNA biosensors. These are portable medical deviceused for diagnosing diseases through specific detection of biomolecules. These DNA SAMsare composed of chemically modified DNA molecules adsorbed onto gold surfaces and aremade by immersing a gold surface in a solution of thiol-modified DNA. Previous literature hasshown that an applied potential to the gold surface during the DNA immersion enhances DNASAM formation by decreasing self-assembly time and forming DNA SAMs of higher quality.These electrodeposited DNA SAMs were observed using average measurement techniqueson the entire surface. These averaged techniques are unable to characterize heterogeneousfeatures in DNA SAMs. Instead, an imaging technique such as in-situ electrochemical fluores-cence microscopy (iSEFMI) can be used to investigate these electrodeposited DNA SAMs.In this work, DNA SAMs were electrodeposited onto single crystal bead electrodes. Theseelectrodes were key for examining DNA SAMs on different surface crystallographies (i.e. sur-face atomic arrangements). Using iSEFMI, the electrodeposited DNA SAMs on the single crys-tal bead electrodes were characterized, with DNA coverage measurements on each surfacecrystallography possible. It was established that applying either negative or positive poten-tials resulted in DNA SAMs of high coverage with positive potentials resulting in more uniformDNA coverages across all surface crystallographies. Applying a modulating potential furthercreated DNA SAMs of slightly higher DNA coverages. The effect of an applied potential onDNA SAMs adsorbing onto different surface crystallographies was also investigated. Both aconstant potential and a modulating square wave potential will be applied and the resultingDNA SAMs studied. The presence of specifically adsorbing anions in solution was found toaffect both the gold surface and the potential-assisted DNA deposition resulting in a DNA SAMof unusual character.iiiAbstractUnderstanding the factors that affect potential-assisted DNA deposition will enable the for-mation of an optimal procedure for manufacturing DNA SAMs. With the appropriate variablescontrolled, DNA SAMs can be tailored for their application in DNA biosensors and eventualuse in point-of-care devices.ivLay SummaryProviding personalized health care has been a proposed vision for the future. This is pos-sible with the use of biosensors, which are portable devices capable of diagnosing diseaseby detecting specific molecules present in a diseased patient. Among the biosensors beingdeveloped, DNA biosensors are one promising type. However, they have been difficult to man-ufacture with high reproducibility with control of its characteristics. Typically DNA biosensorsare made by depositing chemically-modified DNA onto a gold surface. Controlling this processis possible by applying voltages to the gold surface to speed up the DNA deposition process.To observe in detail how the electrical voltage affects the deposition process, the chemically-modified DNA on the gold surfaces are imaged using fluorescence microscopy. This allowsinvestigation into factors which can change the deposition process. This will eventually enablethe determination of which procedure creates a DNA biosensor reproducibly with the mostoptimal sensing capabilities.vPrefaceAll experimental work shown in this thesis was performed by the author at the Advanced Mate-rials and Process Engineering Laboratory (AMPEL) in the Brimacombe building at the Univer-sity of British Columbia (UBC). The work was done under the supervision of Prof. Dan Bizzotto(UBC) and Prof. Hua-Zhong Yu (SFU). Portions of this thesis have been published or are inpending manuscripts.• Data and discussion of the data presented in chapters 5 and 7 have been publishedin a journal article (Leung, K. K.; Gaxiola, A. D.; Yu, H.-Z. & Bizzotto, D. Tailoring theDNA SAM surface density on different surface crystallographic features using potentialassisted thiol exchange Electrochim. Acta, 2018, 261, 188-197). All experiments pre-sented in this publication were performed and analysed by the author of this thesis underthe guidance of Prof. Dan Bizzotto. Andrea Diaz Gaxiola optimized the electrodeposi-tion experimental setup prior to collecting experimental data by the thesis author. Themanuscript was structured, written and edited by Prof. Dan Bizzotto, Prof. Hogan Yu andthe thesis author.• All of contents of chapter 6 and a small section of chapter 5 are included in a journalarticle (Leung, K. K.; Yu, H.-Z. & Bizzotto, D., Electrodepositing DNA Self-AssembledMonolayers on Au: Detailing the Influence of Electrical Potential Perturbation and SurfaceCrystallography ACS Sensors, 2019, 4, 513-520 ). All the experimental data in this workwas performed and analysed by the thesis author under the guidance of Prof. DanBizzotto. The manuscript was structured and written by the thesis author and edited withProf. Dan Bizzotto and Prof. Hogan Yu.• Some of the content in chapter 7 are taken from a manuscript in preparation. The experi-mental work was done by the author of this thesis with supporting data collected by IsaacviPrefaceMartens. The manuscript was written and drafted by the author then edited by Prof. DanBizzotto.viiTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiLay Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiiList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiiNomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxviiDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xxix1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Scope of Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Background Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.1 Electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.1.1 Gold Surface Crystallography . . . . . . . . . . . . . . . . . . . . . . . . 52.1.2 The Metal-Solution Interface and the Electric Double Layer . . . . . . . . 132.1.3 Faradaic Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202.1.4 Electrochemical Measurements . . . . . . . . . . . . . . . . . . . . . . . 22viiiTable of Contents2.2 Self-Assembled Monolayers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302.2.1 Formation of Self-Assembled Monolayers . . . . . . . . . . . . . . . . . . 302.2.2 The Structure of Self Assembled Monolayers . . . . . . . . . . . . . . . 342.2.3 Electrochemical Measurements of SAMs . . . . . . . . . . . . . . . . . . 372.3 Fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402.3.1 Fluorescence Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . 402.3.2 Quenching at Metal Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . 403 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433.1 DNA Self-Assembled Monolayers . . . . . . . . . . . . . . . . . . . . . . . . . . 433.1.1 Deoxyribonucleic acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433.1.2 Making DNA Self-Assembled Monolayers . . . . . . . . . . . . . . . . . . 453.1.3 Applications of DNA SAMs as DNA Biosensors . . . . . . . . . . . . . . 483.1.4 Current Challenges with DNA SAM Application . . . . . . . . . . . . . . . 563.1.5 Potential-Assisted Formation of DNA SAMs . . . . . . . . . . . . . . . . . 593.2 in-situ Electrochemical Fluorescence Microscopy to investigate DNA SAMs . . . 613.2.1 Combining Electrochemistry with Fluorescence Microscopy . . . . . . . . 613.2.2 iSEFMI to Study Alkanethiol Self-Assembled Monolayers on Metal Elec-trodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623.2.3 Observed Surface Crystallography Dependence for Thiol SAMs . . . . . 643.3 Contributions of the Research Presented in this Thesis . . . . . . . . . . . . . . 644 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.1.1 Gold Substrates and Cleaning . . . . . . . . . . . . . . . . . . . . . . . . 664.1.2 Modifying Au substrates . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.1.3 Electrolyte Solutions for Electrochemical Measurements . . . . . . . . . 694.2 Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694.2.1 Instrumentation and Apparatus . . . . . . . . . . . . . . . . . . . . . . . . 694.2.2 in-situ Electrochemical Fluorescence Microscopy . . . . . . . . . . . . . 714.2.3 Electrochemical DNA Coverage Measurements . . . . . . . . . . . . . . 75ixTable of Contents4.2.4 Correlating Fluorescence Microscopy Measurements to ElectrochemicalDNA Coverage Measurements . . . . . . . . . . . . . . . . . . . . . . . 775 Formation of DNA SAMs via Potential-Assisted Thiol-Exchange . . . . . . . . . . 795.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835.3.1 Characterizing the MCH SAM . . . . . . . . . . . . . . . . . . . . . . . . 835.3.2 Thiol Exchange of MCH layers with DNA at OCP . . . . . . . . . . . . . . 845.3.3 Capacitance measurements during DNA thiol exchange with MCH SAM 855.3.4 Fluorescence Characterization of MCH/DNA layers . . . . . . . . . . . . 885.3.5 Thiol-exchange at Various Potentials . . . . . . . . . . . . . . . . . . . . 925.3.6 Assessing the Local Environment of the DNA SAMs made with Potential-Assisted Thiol-Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . 965.3.7 Fluorescence Characterization of DNA SAM Formation on Clean Au(DNA/MCH) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1025.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1046 The Effect of a Square-Wave Modulated Potential during DNA Deposition . . . . 1066.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1066.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1086.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1106.3.1 Potential-Assisted Thiol exchange with Square-Wave Modulated Poten-tial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1106.3.2 Positive Potential Limit and Potential-Assisted DNA Thiol-Exchange . . . 1126.3.3 Square-Wave Modulated Potential-Assisted DNA Adsorption on a CleanAu Bead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1156.3.4 Positive Potential Limit and DNA Deposition onto Clean Au . . . . . . . . 1206.3.5 Charging of the Interface during Potential-Assisted DNA Deposition . . . 1206.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124xTable of Contents7 The Influence of Chloride Anions in the Electrolyte During DNA Deposition . . . 1267.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1267.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1287.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1297.3.1 Influence of Chloride on a Mercaptohexanol SAM . . . . . . . . . . . . . 1297.3.2 Influence of Chloride on a Potential-Assisted DNA Thiol-Exchange . . . . 1307.3.3 Potential-Assisted DNA Adsorption on Clean Au in Chloride . . . . . . . 1327.3.4 Potential-Assisted DNA Adsorption on Clean Au without Chloride . . . . 1347.3.5 Potential-Assisted DNA Adsorption on Clean Au in Phosphate Buffer . . 1387.3.6 Analysis of DNA SAM Local Environment for Different Surface Crystallo-graphies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1417.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1448 Conclusion and Future Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1498.1 Summary of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1498.2 Future Work and Application to DNA Biosensors . . . . . . . . . . . . . . . . . . 1548.2.1 Testing the Hybridisation Efficiency of DNA SAMs made with ControlledCoverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1548.2.2 Potential-Assisted DNA Deposition containing Secondary structures . . . 1558.2.3 Potential-Assisted DNA Deposition on Polycrystalline Surfaces with Im-age Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1558.3 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159AppendicesA T-test Comparison of DNA Coverages . . . . . . . . . . . . . . . . . . . . . . . . . 181B Demonstration on the Noise Propagation during Image Analysis . . . . . . . . . 184C SEM Images of DNA Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186xiList of Tables2.1 The surface crystallographies located between (111) to (100) and (100) to (110)are listed and are known as higher index planes. The step notation for eachsurface crystallography is listed, which helps describe the atomic arrangementfor each. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84.1 Approximate size per region of interest for each crystallographic feature in num-ber of pixels and in μm2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75A.1 Example of calculated T’s for comparison of the Potential-assisted Thiol Ex-change at Edep =0.4V and ESWdep = 0.4V/-0.3V(50 Hz). These conditions had 2replicates. These results correspond to Figure 6.3 . . . . . . . . . . . . . . . . . 182A.2 Example of calculated T’s for comparison of the Potential-assisted DNA Adsorp-tion on bare Au at Edep=0.4V and ESWdep = 0.4V/-0.3V(50 Hz). These conditionshad 2 replicates. These results correspond to Figure 6.8 . . . . . . . . . . . . . . 183xiiList of Figures2.1 Schematic of a face-centered cubic lattice, the unit cell of a gold crystal. . . . . . 62.2 Planes drawn through the face centered cubic lattice (top) from which low indexsurface crystallographies (100), (111) and (110) originate. . . . . . . . . . . . . . 72.3 Atomic model of the spherical single crystal electrode with red atoms indicating(111) surface crystallography, cyan atoms indicating (100) crystallography andgreen atoms indicating (110) crystallography. . . . . . . . . . . . . . . . . . . . . 92.4 A “stereographic” triangle showing the relative orientation of surface crystallo-graphies on a 2-dimensional image of the 3-dimensional spherical surface. Lowindex planes (111) and (110) and high index planes (311) and (210) are shownrelative to the (100). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.5 Atomic models showing the atomic arrangement of high index planes (311) and(210). Red atoms indicate the second layer of atoms underneath surface atoms. 102.6 Charge distribution at the surface due to electron “spill-over” (above) and “smooth-ing” as a result of the roughened features (below). . . . . . . . . . . . . . . . . . 122.7 Schematic depicting three types of metal substrates used: (a) Single crystalelectrode, (b) single crystal bead electrodes and (c) polycrystalline substrates. . 122.8 a) Lateral force microscopy measurements of a gold polycrystalline surface withthe (b) electron back scattering diffraction image of gold surface shown. . . . . . 142.9 Schematic depicting the electric double layer composed of a positive chargedmetal surface, (a) the Helmholtz layer, (b) the diffuse layer and (c) bulk electrolyte. 152.10 The calculated effective thickness of the diffuse layer decreases with increasingelectrolyte concentration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17xiiiList of Figures2.11 (a) The calculated capacitance from the Guoy-Chapman Stern model as a func-tion of potential φ0 for a metal surface in 1 mM to 1 M electrolyte containingmonovalent cations and anions. (b) Differential capacitance measurements of aHg electrode in 1 mM to 1 M NaF electrolyte. . . . . . . . . . . . . . . . . . . . . 182.12 Potential of zero charge (#) varies across the surface crystallographies alongthe zone axis (110) to (111) to (100) to (110). A correlation between the densityof “broken bonds” (+) and the potential of zero charge is evident. . . . . . . . . . 192.13 a) Schematic of a typical 3-electrode electrochemical cell with a working elec-trode (WE), reference electrode (RE) and counter electrode (CE) if necessary.(b) The corresponding simplified circuit model (RC)R depicting faradaic andnon-faradaic processes. (c) If the solution resistance Rs is high, then a counterelectrode (CE) is added alleviating the resistance between the reference andworking electrode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232.14 The potential profile applied for electrochemical measurements: (a) chronoam-perometry (b) cyclic voltammetry and (c) electrochemical impedance spectroscopy.The typical readout for each system is shown below: (a) the measured currentover time, (b) a cyclic voltammogram and (c) a Nyquist plot. . . . . . . . . . . . 232.15 Example chronocoulometric response where a gold electrode with Ru(NH3)3+6adsorbed to the surface is immersed in an electrolyte containing Ru(NH3)3+6 . . . 242.16 Example cyclic voltammogram of reduction and oxidation of Ru(NH3)3+6 /Ru(NH3)2+6(a) adsorbed at the electrode surface and (b) diffusing from solution. . . . . . . . 272.17 Vector diagram indicating the impedance vector Z. . . . . . . . . . . . . . . . . 292.18 Simulated EIS data from a series RC circuit with a 100 Ω resistor and 1 µFcapacitor. The Nyquist plot (left), Bode plot (middle) and Bode phase diagram(c) are outputs from an EIS measurement. . . . . . . . . . . . . . . . . . . . . . 292.19 EIS data simulated from an electrochemical system modeled as a R(C[RW])circuit (a.k.a. Randle’s circuit). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312.20 Schematic of a alkanethiol SAM containing mercaptohexanol on a gold surface. . 322.21 Spatial arrangement of an alkanethiol SAM on Au(111) with (√3 × √3)R30◦organized in c(4x2) lattices (red) or 2√3× 3 lattices (blue). . . . . . . . . . . . . . 34xivList of Figures2.22 Schematic showing the Au-adatom-dithiolate structures which exist in methylth-iolate SAMs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.23 Schematic showing the lattices (√3×√3)R30◦ and c(4× 2) that have been de-tected via STM for long chain alkanethiol containing SAMs. . . . . . . . . . . . . 362.24 CVs showing the reductive desorption of decanethiol SAMs in 0.1 M KOH on (a)polycrystalline gold and (b) monocrystalline surfaces. . . . . . . . . . . . . . . . 392.25 Simulated quantum yield of AlexaFluor488 depending on the distance from ametal surface at distances up to 50 nm. . . . . . . . . . . . . . . . . . . . . . . . 423.1 Depiction of four nucleotides interacting via hydrogen bonding: cytosine (red)with guanine (green), and thymine (yellow) with adenine (blue). . . . . . . . . . . 443.2 A schematic showing hexagonal-packed single stranded DNA (grey circles) onAu (111) for a DNA SAMs made in (a) 0.33 M NaCl and in (b) 1.2 M NaCl. . . . 463.3 Schematic of a molecular beacon used to detect the presence of a target DNAsequence in solution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503.4 Example DNA biosensors approaches which use (a) fluorescence labelling withthe molecular beacon design, (b) potential-induced actuation with measure-ments, (c) redox active Ru(NH3)6Cl3, (d) Ferri/ferrocyanaide, (e) redox activeintercalators, (f)(g)(h) covalent redox active labels. . . . . . . . . . . . . . . . . . 513.5 Hybridisation efficiencies measured using surface plasmon resonance for DNASAMs with varying coverages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.1 a) Projection of the surface of a single crystal bead with certain surface crystallo-graphic features labeled. b) Brightfield image of a single crystal bead electrodewith the same surface crystallographies. . . . . . . . . . . . . . . . . . . . . . . . 674.2 Cyclic voltammogram of a Au spherical electrode in approximately 0.1 M H2SO4cycled between -0.40 V and 1.45 V vs. SCE at 50 mV/s scan rate . . . . . . . . 684.3 Typical electrochemical apparatus used for substrate cleaning and electrochem-ical characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70xvList of Figures4.4 For electrochemical fluorescence microscopy, the Au substrate in a spectro-electrochemical cell with a Pt coil electrode and a teflon stopcock salt bridgecontaining a saturated calomel electrode. The cell is mounted above an epi-fluorescence microscope where fluorescence images are taken. . . . . . . . . . 714.5 (a) Schematic of the filter cube that separates excitation and emission in theinverted epi-fluorescence set up. (b) The excitation (blue) and emission (green)spectra for the fluorophore AlexaFluor®488 is shown as well as the transmissionspectra for the excitation filter, the emission filter and dichroic mirror used in thefluorescence microscopy configuration. . . . . . . . . . . . . . . . . . . . . . . . 724.6 The alternating potential profile applied during iSEFMI. At each potential step afluorescence image was taken. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 724.7 Brightfield image of (a) a spherical single crystal bead electrode where (b) theAu(111) facets and (c) Au(100) facets are located. These are indicated by flatregions surrounded by terraces. . . . . . . . . . . . . . . . . . . . . . . . . . . . 744.8 a) Brightfield image of single crystal bead electrode with green ellipsoids indi-cating regions of interest of different surface crystallographic features. b) Thefluorescence image of a DNA SAM assembled on the same electrode with twostereographic triangles containing the crystallographic features labelled. . . . . . 744.9 Cyclic voltammograms obtained from the first scan between -0.45V and 0.05 Vvs. SCE at scan rate 0.5 V/s for individual DNA SAMs on different Au singlecrystal bead electrodes immersed in a) 2.5 μM and b) 5 μM of RuHex. . . . . . . 764.10 For chronocoulometric measurements,(a) a potential pulse from 0V to -0.45 Vis applied, (b) the current measured when applying -0.45 V is integrated (grey)to give the (c) charge and plotted against square root time. . . . . . . . . . . . . 77xviList of Figures4.11 (a) Fluorescence image of an example (relatively) homogeneous DNA SAM(made with immersion for 30 min in MCH, 1 hour in DNA at Edep=0.4 V/SCE) ona single crystal Au bead electrode. The region of interest, where the averagefluorescence intensity of the entire electrode was measured, is shown as a blackpolygon. Select crystallographic features are labeled to indicate the symmetryof the entire electrode with respect to the region of interest. (b) The relationbetween this average fluorescence intensity and DNA coverage measured withelectrochemical DNA coverage measurements in the presence of RuHex is shown. 785.1 a) Schematic of the formation of a DNA SAMs using potential-assisted DNAthiol-exchange. b) Depiction of the 2-electrode apparatus used for potential-assisted DNA deposition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 825.2 Potential profile over time during application of (a)Edep where the OCP is mea-sured followed by an EIS measurement then Edep is applied while taking mul-tiple EIS measurements. Edep=+0.4V/SCE is shown here as an example. (b)For depositions at OCP, the OCP is measured for some time and then a EISmeasurement is taken at the last measured OCP. . . . . . . . . . . . . . . . . . 835.3 Capacitance measured for a MCH SAM coated gold bead electrode in immo-bilization buffer. The line shows the capacitance measured from scanning thepotential (with 5 mV rms 200 Hz perturbation ) from 0 V to 0.6 V or 0 to -0.6 Vdone using a three electrode setup. . . . . . . . . . . . . . . . . . . . . . . . . . . 855.4 Distribution of the initial open circuit potential measured for MCH SAMs uponimmersion in the immobilization buffer containing DNA . . . . . . . . . . . . . . . 865.5 The open circuit potential (OCP) of MCH SAM covered gold bead electrodeswhen immersed in the DNA deposition solution for various deposition times (5to 60 min). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 865.6 Capacitance measurements made from impedance spectroscopy during DNAdeposition at various Edep. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87xviiList of Figures5.7 Fluorescence images taken of MCH/DNA layers prepared a) at OCP (no appliedpotential) and b) at Edep= +0.40 V/SCE. Images from left to right correspond toincreasing time in the deposition solution. Each image is from a different elec-trode resulting in a different orientation. The stereographic triangle and crys-tallographic regions analyzed are shown on the images. All images are falsecoloured to represent intensity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 895.8 Comparison of the fluorescence images (left) and fluorescence intensity (right)measured from selected regions on the gold bead electrode after 60 min thiol-exchange process performed at OCP with the application of the OCP during EISmeasurements to that with only monitoring the OCP. . . . . . . . . . . . . . . . . 895.9 Fluorescence intensity, which correlates to the DNA coverage, measured acrossdifferent DNA immersion times, measured on the low-index regions (111), (100),and (110), for MCH/DNA layers made at OCP (open points) and layers made atEdep= +0.40 V vs SCE (filled points). . . . . . . . . . . . . . . . . . . . . . . . . . 905.10 Fluorescence intensities measured from MCH/DNA on gold bead electrodesprepared via thiol-exchange at Edep= +0.40 V vs. SCE (filled points) and atOCP (open points) for the high index regions (311) and (210). . . . . . . . . . . . 905.11 Fluorescence images taken of MCH/DNA layers made with at 1 h DNA immer-sion times at (a)Edep= +0.50 V, (b)Edep= -0.45 V and (c)Edep= -0.55 V. . . . . . . 935.12 Fluorescence intensities measured from various crystallographic regions for thiol-exchanged MCH/DNA SAMs prepared for 1 h at different applied (or OCP) po-tentials as raw intensities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945.13 Example calculation of the fluorescence modulation of a particular region of in-terest on a DNA SAM. Images taken during iSEFMI of the (111) at -0.4V (Fmax)and +0.35 V (Fmin) are shown. Images showing the subsequent calculations forFmax − Fmin and Fmod are displayed. . . . . . . . . . . . . . . . . . . . . . . . . 97xviiiList of Figures5.14 Models (a and b) of the extent of DNA angular mobility during potential drivenDNA reorientation with a cylindrical framework for the DNA (green) and a fluo-rophore at the distal end (red circle). A modeled estimate of the fluorescencemodulation for both types of DNA reorientation as a function of c) DNA surfacecoverage or d) the estimated fluorescence intensity from iSEFMI measurements. 995.15 Analysis of the ability of the DNA SAM to undergo potential driven reorientation(resulting in a modulation of fluorescence intensity, Fmod) for a range of surfaceconcentrations. Fluorescence Modulation plotted against fluorescence intensity(bottom x axis) and DNA coverage (top x axis) for all regions of interest mea-sured, displaying negative correlation between DNA coverage and DNA localcrowdedness (- - -) until a maximum coverage is attained with no detectablefluorescence modulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1005.16 The magnitude of the fluorescence modulation for DNA SAMs for the five crys-tallographic regions studied as a function of the fluorescence intensity mea-sured. The Figure includes DNA SAMs made via thiol-exchange at OCP (openpoints) and Edep=0.4 V/SCE (filled points). . . . . . . . . . . . . . . . . . . . . . . 1015.17 Fluorescence Images of DNA SAMs made with 5 minute DNA adsorption atvarious potentials a)OCP b) Edep= +0.4 V c) b)Edep= +0.5 V . . . . . . . . . . . . 1035.18 Bar graph showing average fluorescence intensities of DNA SAMs on certainsurface crystallographies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1036.1 Schematic depicting electrodeposition set up for 2 electrode system used in theprevious chapter (a) and the 3 electrode system used to apply ESWdep (b) . . . . . 1096.2 Potential profile applied during DNA deposition at ESWdep (50 Hz square wave,+0.4V to -0.3V/ SCE) applied during the DNA immersion time (5 minutes). . . . 1096.3 Fluorescence images obtained through iSEFMI of DNA SAMs prepared usingpotential-assisted DNA thiol-exchange (for 5 min) at (a) Edep=+0.4V/SCE and(b)ESWdep (50 Hz square wave, +0.4V to -0.3V/ SCE). . . . . . . . . . . . . . . . . 111xixList of Figures6.4 Fluorescence images of DNA SAMs prepared via DNA thiol-exchange (for 5min) at a) Edep=+0.4V/SCE and (b) ESWdep ( +0.4V to -0.3V/SCE, 50 Hz) for vary-ing concentrations of DNA in the deposition solution. . . . . . . . . . . . . . . . . 1126.5 Average fluorescence intensity for five selected regions representing low indexplanes (111), (100), (110) and high index planes (210), (311) on DNA SAMsmade at Edep=+0.4 V/SCE (filled points) and ESWdep = +0.4 V to -0.3 V/SCE (50Hz)(closed points). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1136.6 Fluorescence images obtained through iSEFMI of DNA SAMs prepared withDNA thiol-exchange (for 5 min) at a) Edep=+0.5V/SCE and (b) ESWdep ( +0.5Vto -0.3V/ SCE, 50 Hz) with a stereographic triangle to label five surface crys-tallographies. c) measured fluorescence intensities for these crystallographicsurfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1156.7 Comparison of fluorescence intensity for DNA SAM thiol exchange for five se-lected regions prepared at OCP, Edep (+0.4V or +0.5V) and ESWdep (+0.4 V to -0.3V/SCE (50Hz) or +0.5 V to -0.3 V/SCE (50Hz)). . . . . . . . . . . . . . . . . . . . 1166.8 Fluorescence images of DNA SAMs prepared on clean Au (for 5 min) at a)OCP; b)Edep =+0.4 V/SCE; c) ESWdep = +0.4 V to -0.3 V/SCE (50Hz) followed byimmersion in 1mM MCH in IB for 90 min.; d) Measured fluorescence intensitiesfor five selected DNA SAMs on these surface crystallographies. . . . . . . . . . . 1176.9 Fluorescence images of DNA SAMs made with DNA adsorption onto a cleanAu bead electrode surface (for 5 min) using a) Edep =+0.4 V/SCE and (b) ESWdep= +0.4 V to -0.3 V/SCE (50Hz) with increasing concentrations of DNA in thedeposition buffer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1186.10 Measured fluorescence intensities on the low index surfaces (111), (100) and(110) and high index planes (311) and (210) for DNA SAMs at varying DNA con-centrations with 5 minute DNA adsorption at Edep=+0.4V/SCE (closed points,solid line) and ESWdep = +0.4 V to -0.3 V/SCE (50Hz) (open points, dotted line). . . 119xxList of Figures6.11 Fluorescence images of DNA SAMs prepared on clean Au (for 5 min) at a)OCP; b)Edep =+0.4 V/SCE; c) ESWdep = +0.4 V to -0.3 V/SCE (50Hz); d) Edep=+0.5 V/SCE; e) ESWdep = +0.5 V to -0.3 V/SCE (50Hz), followed by immersion in1mM MCH in IB for 90 min. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1216.12 Measured fluorescence intensities for DNA SAM manufactured via adsorptionon clean Au for five selected surface crystallographies prepared at OCP, Edep(+0.4V or +0.5V) and ESWdep (+0.4 V to -0.3 V/SCE (50Hz) or +0.5 V to -0.3V/SCE (50Hz)). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1216.13 Current measured over 0.2 seconds at the start and end of the potential-assisteda) DNA Thiol-exchange and b) DNA adsorption on Bare Au while applying anESWdep = +0.4 V to -0.3 V/SCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1236.14 Charge measured over 0.2 seconds at the start and end of potential-assisteda) DNA Thiol-exchange and b) DNA adsorption on clean Au while applying anESWdep = +0.4 V to -0.3 V/SCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1236.15 Magnitude of charge passed to the gold bead electrode interface during the first0.2s of the 5 min electrodeposition using ESWdep for DNA SAMs created using a)thiol-exchange method or b) DNA adsorption onto clean gold followed by MCH. . 1247.1 Capacitance of a MCH SAM on the gold bead electrodes measured starting at0V and (a) scanning to negative potentials or (b) scanning to positive potentialsrevealing the influence of Cl– in the stability of the MCH SAM. . . . . . . . . . . 1307.2 a) Fluorescence images of thiol-exchanged MCH/DNA SAMs prepared by elec-trodeposition at Edep= +0.40 V for 1 hour in 1 μM DNA diluted in immobilizationbuffers with a large [Cl– ] or a significantly lower [Cl– ].Image exposure time was3s with EM gain 50 and falsely coloured to emphasize fluorescence intensities.b) Fluorescence intensities on select surface crystallographic regions are shown. 131xxiList of Figures7.3 Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdep for5 minutes in a 0.5 μM DNA solution in TRIS IB containing Cl– . During iSEFMImeasurements, a modulated potential profile was applied with a fluorescenceimage taken at each step. The fluorescence intensity on (111) and (100) duringthis measurement is shown. (c) The measured fluorescence modulation plottedover the maximum fluorescence intensity for DNA SAMs on all other surfacesmade at different [DNA] (0.1 μM to 1 μM) while applying Edep or ESWdep for 5 minutes.1357.4 a) Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdepfor 5 minutes in a 0.5 μM DNA solution in TRIS IB containing NO –3 . The fluo-rescence intensity was measured on (111) and (100) during the iSEFMI mea-surements. The changes in the fluorescence intensity as a result of DNA re-orientation are smaller for the DNA SAMs made while applying ESWdep . (c) Themeasured fluorescence modulation plotted over the maximum fluorescence in-tensity for DNA SAMs on all other surfaces made at different [DNA] (0.25 μM to0.5 μM ) while applying Edep or ESWdep for 5 minutes. . . . . . . . . . . . . . . . . . 1367.5 Schematic depicting the formation of high surface density clusters for two DNASAMs with the same DNA coverage. The top schematic depicts the DNA SAMmade while applying Edep results in homogeneous non-clustered DNA SAMsand the bottom depicts the DNA SAM made while applying ESWdep containinghigh density clusters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1377.6 Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdepfor 5 minutes in a 0.5 μM DNA solution in a phosphate IB containing Na2SO4.The fluorescence intensity was measured on (111) and (100) during the iSEFMImeasurements. The fluorescence modulations are smaller for the DNA SAMsmade while applying ESWdep suggesting cluster formation. (c) The measured fluo-rescence modulation plotted over the maximum fluorescence intensity for DNASAMs on all other surfaces made at different [DNA] (0.1 μM to 1 μM) while ap-plying Edep or ESWdep for 5 minutes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 140xxiiList of Figures7.7 Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdepfor 5 minutes in a 0.5 μM DNA solution in a phosphate IB containing NaCl.The fluorescence intensity was measured on (111) and (100) during the iSEFMImeasurements. The reorientation of the DNA , measured during iSEFMI, isnot restricted for the DNA SAMs made while applying ESWdep . (c) The measuredfluorescence modulation plotted over the maximum fluorescence intensity forDNA SAMs on all other surfaces made at different [DNA] (0.25 μM to 0.5 μM)while applying Edep or ESWdep for 5 minutes. . . . . . . . . . . . . . . . . . . . . . . 1427.8 Fluorescence modulation vs. fluorescence intensity plot with shaded areaswhere the measured fluorescence modulation deviates from the experimentallydetermined maximum (dashed line). . . . . . . . . . . . . . . . . . . . . . . . . . 1447.9 The percent deviation from the ideal behaviour of a homogeneously spacedDNA SAM for DNA SAMs made while applying Edep and ESWdep in their respec-tive IB. The percent deviation represents the portion of the DNA SAM which iscomposed of high density DNA clusters. . . . . . . . . . . . . . . . . . . . . . . 1457.10 Fluorescence modulation vs. fluorescence intensity plots for DNA SAMs on theirrespective crystallographic surfaces: (111), (100), (110), (210) and (311). . . . . 146B.1 (Top row) Demonstration of the effect of noise propagation when calculating thefluorescence modulation ((Fmax − Fmin)/Fmax) of a particular region of intereston a DNA SAM. (Bottom row) Images calculated after the initial images weredespeckled and Gaussian blurred results in reduction of this noise propogation. 184B.2 A comparison of the histogram for the intensities from the calculated fluores-cence modulation ((Fmax − Fmin)/Fmax) image in the previous figure. Withoutany despeckling and Gaussian blurring of the initial images, the histogram ofresulting intensities has a wide standard deviation (above). Additionally, somepixels are lost as the calculation of noisy pixels results in NaN. As a result ofthe despeckling and Gaussian blurring of the initial images, the histogram (bot-tom) of resulting intensity values has a lower standard deviation. Only valuesbetween 0 and 1 are shown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185xxiiiList of FiguresC.1 SEM images of DNA SAMs on Au(111) manufactured at Edep =+0.4 V/SCE; c)ESWdep = +0.4 V to -0.3 V/SCE (50Hz)l . . . . . . . . . . . . . . . . . . . . . . . . . 186xxivNomenclatureΓ Surface CoverageΓDNA DNA Surface CoverageA Surface Area of ElectrodeAp Area of Peak in cyclic voltammogramC CapacitanceCC ChronocoulometryCE Counter ElectrodeCV Cyclic VoltammetryDNA Deoxyribonucleic AciddsDNA Double-stranded DNAEdep Constant Deposition PotentialEdep Deposition PotentialESWdep Square Wave Deposition PotentialEIS Electrochemical Impedance SpectroscopyF Faraday’s ConstantFl. Int. Fluorescence IntensityHPLC High-Performance Liquid ChromatographyxxvNomenclatureIB Immobilization BufferiSEFMI in-situ electrochemical fluorescence microscopy imagingMCH MercaptohexanolMeOH MethanolNA Numerical ApertureOCP Open Circuit PotentialOHP Outer Helmholtz planepzc Potential of Zero ChargeQ ChargeQY Quantum YieldRE Reference ElectrodeRNA Ribonucleic AcidROI Region of InterestRuHex Ruthenium HexaammineSAM Self-Assembled MonolayerSCE Saturated Calomel ElectrodessDNA single-stranded DNASTM Scanning Tunnelling MicroscopyTCEP Tris(2-Carboxyethyl)Phosphine-HydrochlorideTRIS TrisaminomethaneWE Working ElectrodexxviAcknowledgementsI would like to thank so many people who were part of my journey in grad school:My two supervisors Prof. Dan Bizzotto and Prof. Hua-Zhong (Hogan) Yu, I would like tothank the most. I learned so much with their guidance and support through the years I haveworked with them. I may not have always been the brightest student, but I am glad that both ofthem tolerated me and my antics. I have learned a great deal from the both of them, especiallywhen it came to due diligence in research work and exercising patience.I have been very appreciative of my labmates since my time in grad school: Jannu Casanova-Moreno, Santa-Maria Gorbunova, Elizabeth Fisher, Zhinan (Landis) Yu, Tony Yang, IsaacMartens, Rochita Sundar, Alexandra Verhaven, Tianxiao Ma. I will always treasure the in-formative, funny and sometimes weird discussions that happened in the lab. The advice fromthe labmates who were senior to me were really helpful in my journey through grad school.To my colleagues in Chemistry who worked in the glass-blowing shop and mechanicalshop, thank you for helping me build the material necessary to my project or fixing importantequipment in a time which allowed my research to continue. To colleagues in chemistry infinances and administration, I would like to extend my gratitude to them for dealing with the pa-perwork and explaining the policies so that I could have a smooth time within the department.I have too many friends outside the lab that I would like to thank. To the other grad studentsin chemistry and/or studying at UBC, I am so grateful to have met so many and whom I wasable to connect with on a personal level. Understanding that we are all in the same grind hasmade grad school life a little more bearable. To all my other friends I apologize for being sobusy all the time with grad school and other things. I write this to let you all know that thoughour time spent together may have been short, I won’t forget how meaningful our relationshipswere and how pertinent that it was for my functioning in grad school.I would also like to thank members in a certain university club. Though we may havexxviiAcknowledgementsparted ways due to differences, I am very appreciative of the environment provided where Icould exercise and maintain fitness. I will not forget the sense of comradery we shared whilecompeting and while living our day to day student lives.Finally, I would like to thank my family: Mom and Dad and my big sister Kristel. To mymom and dad, both of you have worked hard in the past so that your daughters could have aproper education. I hope that I have not left your good efforts to waste. Thank you for havingfaith in me, for allowing me to pursue something that I am passionate about and for helping menavigate through the times where I felt lost. To Kristel, you have been my role idol who I lookup to in academics and in life. I know I am not as smart as you but seeing your goal-drivenmentality has motivated me towards chasing my dreams.I would not have been able to be where I am now and I would not be the person who Iam today had it not been for these people. This thesis would not have existed without all yoursupport and with that, I would like to thank you all with an infinite amount of thank yous.xxviiiDedicationTo Sio Yee Leung and Yau-Foon Chan.xxixChapter 1Introduction1.1 MotivationDiagnosing and subsequently monitoring disease progression during treatment depends onthe detection of biomarkers (small molecules, peptides, DNA, RNA, lipids etc.) found in the hu-man body or produced by pathogens or tumours. The low concentrations of these biomarkersin bodily fluids (i.e. blood, urine, saliva) poses challenges for their detection, leading to a re-liance on amplification methods and sensitive instrumentation specific to medical laboratories.Currently, most analysis of these biomarkers can take days. This is particularly problematicfor time-sensitive medical conditions, where delayed treatment leads to disease progressionor infectious spread. Due to these reasons, it is imperative that detection of these biomarkersbe made more affordable and portable via point-of-care devices available outside the confinesof clinical laboratories.1Portable point-of-care devices utilise bioreceptors that recognize biomarkers of interestthrough specific binding. These bioreceptors include but are not limited to antibodies, en-zymes, proteins/peptides and DNA. One proposed type of point-of-care device, the elec-trochemical biosensor, consists of bioreceptors immobilized onto conductive materials suchas carbon, platinum, gold and indium-titanium oxide , have shown great promise in recentyears.2–4 A clear advantage is the simple instrumentation required for these electrochemicalbiosensors compared to optical instrumentation, enabling low cost manufacturing and allowingfacile miniaturization of the technology. Additionally, electrochemical biosensors can be easilyimplemented with the flow-through technology in microfluidic devices which increases the sen-sitivity of the sensor by increased likelihood of biomarkers being detected on small surfaces.2However, there is a clear lack of commercial implementation of electrochemical biosensors11.1. Motivationwith exception to electrochemical glucose sensors.Utilising deoxyribonucleic acid (DNA) oligomers as bioreceptors has become more prefer-able to antibodies which are currently used in clinical assays. Certain assays using DNAas a bioreceptor have been shown to be more sensitive when compared to antibody basedassays.5 Furthermore, DNA has the added advantage of being easy to modify and can besynthesized in-vitro. With the smaller size of DNA compared to antibodies, higher densities ofbioreceptors can be immobilized onto a surface, making for more sensitive biosensors. DNAsensors, consisting of single stranded DNA immobilized to the surface, function by detect-ing complementary strands through hybridisation. DNA sensors, have proven to be versatilewith detection of not only DNA strands with a specific sequence but also to ribonucleic acid(RNA), small molecules, peptides and proteins,6,7 providing alternatives for current biomarkerdetection methods.One of the ways to immobilize DNA molecules onto a surface is by the addition of analkanethiol moiety to the DNA molecules. This thiol covalently interacts with the gold surface,which enables the attachment of DNA onto the surface. The use of gold as a metal substratefor DNA self-assembly opposed to other metals such as silver and copper, is beneficial dueto its abundance, resistance to oxidation and biocompatibility. The alkanethiol adsorption ontogold surfaces had been extensively studied to understand the formation of alkanethiol self-assembled monolayers (SAMs).8 This research had been applied to adsorbing DNA onto goldto form DNA SAMs. Several electrochemical DNA sensors based on DNA SAMs have beenproposed for the detection of DNA strands, proteins, small molecules and ions.7,9 However,their commercialization has been reported to be challenging due to difficulties in detecting lowconcentrations of DNA (as low as 10-18 M2) and other biomarkers (e.g. peptides, proteins,hormones, small molecules) from biological media. With the current sensitivity and detectionlimit of DNA sensors, the use of amplification techniques (e.g. polymerase chain reaction) arenecessary but increases the time between sample intake and detection. Thus, finding methodsto increase the sensitivity of DNA sensors are necessary in order for them to compete with thecurrent biomarker detection methods.10Although much research has been done on alkanethiol SAMs,8,11 forming DNA SAMs ismore complicated due to problems such as aggregation and nonspecific adsorption of DNA.1221.2. Scope of ThesisAt present, irreproducible preparation of such DNA SAMs is prominent with reports of incon-sistent behaviour of some DNA hybridisation sensors.13 This suggests that current assemblyprotocols are not sufficient for controlling the formation of DNA SAMs, with control of the DNAcoverage needed in order to improve sensor performance.Applying an electric potential during alkanethiol SAM formation has been reported in theliterature to result in a stable SAM containing comparatively less defects than those made with-out an applied potential.14 Application of this electric potential to the gold surface during theadsorption of thiol modified DNA is a proposed method to better control DNA SAM formation.15To elucidate the factors that decrease the inconsistencies and nonideal behaviour in the DNASAMs and would enhance their sensitivity as electrochemical DNA biosensors, understandingthe interaction between DNA molecules and the gold surface is crucial. In-situ electrochemi-cal fluorescence microscopy imaging (iSEFMI), a spectroelectrochemical technique previouslyused to study the nature of organic monolayers on metal surfaces, enables comprehensiveexamination of DNA SAMs made using a variety of assembly procedures.16 In this thesis,potential-assisted deposition (or electrodeposition) of DNA onto gold at different conditionsand the resulting DNA SAMs will be analysed with iSEFMI, to create a DNA biosensor withhigher sensitivity.1.2 Scope of ThesisPotential-assisted DNA deposition onto gold to form DNA SAMs was investigated and the re-sulting DNA SAMs characterized with iSEFMI. Fundamental concepts required to understandthe influence of the gold surface atomic arrangement, the influence of the applied potentialto the surface, SAMs and the characterization methods used are detailed in the next chapter.Chapter 3 reviews the literature on DNA SAMs and their application as DNA biosensors as wellas the development of iSEFMI to characterize SAMs. A summary of contributions for the workpresented in this thesis within the context of the literature are also included in this chapter.The experimental chapters consist of results and discussion regarding two DNA SAM man-ufacturing methods have been modified for electrodeposition: DNA adsorption onto a cleangold surface and thiol-exchange of an alkanethiol layer with DNA. The former method is more31.2. Scope of Thesiscommonly used while the latter method is an alternative that will enable better understandingof how the electric potential affects the SAM formation. Additionally, the use of a modulatedelectric potential is compared to the deposition at a constant applied potential. The compo-sition of electrolyte is shown to play a role during electrodeposition, altering the nature of theDNA SAM. These experiments will be detailed in three experimental chapters:• Chapter 5, titled “Formation of DNA SAMs via Potential-Assisted Thiol-Exchange”, in-volves characterization of DNA SAMs made via thiol-exchange of an alkanethiol layerwith DNA while applying a constant potential. The effect of potential on the DNA SAMson surfaces with different surface crystallographic features (i.e. atomic roughness) willbe explored. A direct comparison with potential-assisted DNA adsorption on clean goldwhile applying a constant potential will also be presented.• Chapter 6, titled “The Effect of a Square-Wave Modulated Potential during DNA Depo-sition”, will examine both potential-assisted thiol-exchange and potential-assisted DNAadsorption on clean gold performed while applying a modulated potential and comparedto DNA SAMs made while applying a constant potential. Differences in the DNA SAMsacross different surface crystallographic features will also be shown in this chapter with aproposed mechanism for the enhanced DNA adsorption as a result of potential-assisteddeposition.• Chapter 7, titled “The Influence of Chloride Anions in the Electrolyte During DNA De-position”, will show that performing the DNA deposition in a different electrolyte solutioninduces changes in the DNA SAM. The presence of specifically adsorbing anions willbe shown to result in differences in the DNA coverage and the DNA local environment.DNA SAMs made in either phosphate and TRIS buffers, which are common buffers usedduring DNA deposition, will also be characterized.After exploring these electrodeposited DNA SAMs, future studies are outlined detailing theapplication of electrodeposited DNA SAMs as DNA biosensors will be discussed in the con-cluding chapter.4Chapter 2Background TheoryThis chapter reviews important background concepts and theory required for studying potential-assisted DNA SAM formation on gold. Electrochemistry will be addressed first, with sectionson the metal surface, the solution-metal interface and electrochemical surface characterizationtechniques. Alkanethiol self-assembled monolayers will follow, detailing how self-assembledmonolayers form, their structure and how they are electrochemically characterized. Finally,fluorescence, which is prominently used to characterize the DNA SAMs in this work, will bedescribed.2.1 ElectrochemistryThe majority of the work in this thesis involves applying potential to a metal surface during ad-sorption and electrochemical characterization. The arrangement of atoms at the metal surfaceplays a key role in its electrochemical behaviour and will be shown later to affect the self-assembled monolayer structure. To better understand the effect of electrolyte composition,the electric double layer, which includes the ion distribution at the metal-solution interface asa function of potential at the metal, will be described. Measurement of redox active species insolution and on the metal surface using different voltammetric techniques will be summarizedlast.2.1.1 Gold Surface CrystallographyThough metal surfaces are considerably flat at the macroscopic level, the arrangement of theatoms at the surface influences the adsorption of small molecules. As all of the work in thisthesis involves assembling organic molecules onto gold, only gold surfaces will be discussed.52.1. ElectrochemistryFigure 2.1: Schematic of a face-centered cubic lattice, the unit cell of a gold crystal. It shouldbe noted that the atom sizes are not drawn to scale to clearly show the arrangement of atomswithin the lattice.The surface atomic arrangement, i.e. the surface crystallography, is dictated by the crystalstructure of bulk gold.The unit cell of a gold crystal lattice is a face-centered cube with an edgelength equal to 4.08 Å, containing atoms in the 8 vertices and an atom in the center of each ofthe 6 faces (Figure 2.1). A point in the lattice can be defined using the vector r, which throughthe following equation is the sum of lattice vectors: a, b, c with integers: u, v, w.r = ua+ vb+ wc (2.1)The crystallography of the surface can be defined by drawing a plane through the lattice.The atom arrangement that is apparent along this plane indicates the surface crystallography.(hkl) is used to specify the plane with the Miller indices h,k and l indicating the direction ofthis plane. To explain this notation, we must understand that the plane intersects throughthe lattice vectors a, b and c. Precisely, the plane will intersect along multiples of these unitvectors: (1/h)a, (1/k)b and (1/l)c.As an example, the (100) which is a plane that is parallel tob and c (Figure 2.2 a), intersects at 1a but never intersects b and c (denoted as intersecting∞b and ∞c). Therefore the miller indices are h=1/1=1, k=1/∞=0 and l = 1/∞=0. Regardingother surface crystallographies, the (111) intersects at 1a , 1b, and 1c (Figure 2.2b) and (110)intersects at 1a, 1b and never intersects c (Figure 2.2c).17The surface crystallography, defined by these planes, have a specific arrangement of atoms62.1. ElectrochemistryFigure 2.2: Planes drawn through the face centered cubic lattice (top) from which low indexsurface crystallographies (100), (111) and (110) originate. Atoms in the lattice are not drawnto scale for structure clarity. The atomic models of the low index surface crystallographies(bottom) show the atomic arrangement at the surface. Red atoms indicate a second layer ofatoms underneath surface atoms. Models were created through Crystal Maker.(Figure 2.2 bottom). The atoms on Au (100) are closely packed and have square lattice geom-etry. The (111) contains atoms that are hexagonal close-packed and has the highest density ofatoms among all other surfaces crystallographies. The (110) consist of rows of atoms on top ofunderlying layers of atoms and has rectangular geometry. The three planes, (100), (111) and(110), containing close-packed atoms, are known collectively as the low index planes. Sincethe unit-cell lattice is cubic, the (100) is symmetrically equivalent or has the same atomic ar-rangement as the (010) and (001). These planes can be grouped in a family of planes {100}with {hkl} indicating all symmetrically equivalent planes. Though it is technically incorrect todo so and for simplicity, it should be noted that throughout the thesis, (hkl) is routinely usedinstead of {hkl} to indicate a family of planes.If we were to imagine a lattice containing face-centered cubic unit cells and truncated intoa sphere, the surface would be composed of many different surface crystallographies. Thisconstruct is the basis for the spherical single crystal bead electrode that is prominently usedin this thesis. The orientation of crystallographies at the spherical surface is dictated by the72.1. Electrochemistry100 to 111 Step Notation 100 to 110 Step Notation100 100911 5(100) × (111) 610 6(100) × (110)711 4(100) × (111) 510 5(100) × (110)511 3(100) × (111) 410 4(100) × (110)311 2(100) × (111) or 2(111)×(100)310 3(100) × (110)211 3(111) × (100) 210 2(100) × 110 or 2(110) ×(100)533 4(111) × (100) 320 3(110) × (100)322 5(111) × (100) 430 4(110) × (100)755 6(111) × (100) 540 5(110) × (100)111 110Table 2.1: The surface crystallographies located between (111) to (100) and (100) to (110) arelisted and are known as higher index planes. The step notation for each surface crystallographyis listed, which helps describe the atomic arrangement for each. Adapted with permission fromYu and colleagues.19 Copyright (2015) American Chemical Society.the unit cell in the center of this sphere. One can imagine drawing a plane through the unitcell then projecting this plane to the surface. As a result, along the spherical surface, the(100) and (110) are oriented at a 45 degree angle with respect to each other and the (111)is at 45 degrees with respect to the (110) (Figure 2.3). The relative orientation of the (111),(100) and (110) projected from a 3D surface to 2D image can be labeled using what is knownas a “stereographic” triangle (Figure 2.4). This “stereographic” triangle is used to highlightthe positions of the different surface crystallographies in optical images of the spherical singlecrystal electrodes surface. The regions between (111) to (100), (100) to (110) and (110) to(111), are known as crystallographic zone axes. Along each region, the surface crystallographysystematically changes from the atomic arrangement of one low index plane to that of anotherlow index plane. These surface crystallographies, collectively known as high-index planes,consist of atomically rougher surfaces with steps and terraces. Step notation can be usedto describe the atomic arrangement of these high-index surface crystallographies in terms ofterraces with low index-planes and steps of a different lower index plane (Table 2.1).18 Forexample, (911) is described to have terraces with (100) surface crystallography 5 atoms wideand a step with (111) surface crystallography that is one atom wide.In this thesis, two high index planes of interest are (311) and (210), with the former located82.1. ElectrochemistryFigure 2.3: Atomic model of the spherical single crystal electrode with red atoms indicating(111) surface crystallography, cyan atoms indicating (100) crystallography and green atomsindicating (110) crystallography. The number of atoms in this model is significantly smallerthan that of a typical macro sized single crystal to illustrate the atomic arrangements on themodel.between the (111) and (100) and the latter between the (100) and (110) (Figure 2.4). The (311)and (210) are called turning points and are the most atomically rough surface crystallographywithin their respective crystallographic zone axis. The (311) and (210) can each be describedin two ways using step notation. The (311) is composed of (111) terraces that 2 atoms widewith (100) steps and is identical to (100) terraces that are 2 atoms wide with (111) steps (Figure2.5). The (210) is made of (110) terraces of 2 atom width with (100) steps and is identical (100)terraces 2 atoms wide with (110) steps (Figure 2.5).18The arrangement of atoms in the different surface crystallographies has known effects onthe surface energy. Atoms in the bulk metal are stabilized by nearest neighbour interactionswith surrounding atoms, as compared to surface atoms. Depending on the surface crystal-lography, a decrease in nearest neighbour interactions occurs when creating a surface fromthe bulk metal. This decrease in nearest-neighbour interactions is quantified by the “densityof broken bonds”.20,21 The greater the density of broken bonds, the higher the surface energy.Among the low index planes, the density of broken bonds and the surface energy increasesfrom (111) to (100) to (110). As the higher index planes located along the zone axes are92.1. ElectrochemistryFigure 2.4: A “stereographic” triangle showing the relative orientation of surface crystallogra-phies on a 2-dimensional image of the 3-dimensional spherical surface. Low index planes(111) and (110) and high index planes (311) and (210) are shown relative to the (100).Figure 2.5: Atomic models showing the atomic arrangement of high index planes (311) and(210). Red atoms indicate the second layer of atoms underneath surface atoms. Models werecreated through Crystal Maker.102.1. Electrochemistrymore atomically rough, the density of broken bonds is generally higher. The (210), locatedbetween (110) and (100) has the highest density of broken bonds and therefore the highestsurface energy. The (311) has a higher density of broken bonds than (111) and (100) but alsohas a similar density of broken bonds to the (110) and therefore similar surface energy.21 Thesurface energy plays a role in the preferential formation of these surfaces. Surface crystallo-graphies of lower surface energy have a tendency to form over surface crystallographies withhigher surface energy. Certain surfaces have also been seen to undergo reconstruction wherethe atoms rearrange into a configuration to lower the surface energy.22The surface crystallography and its associated surface energy affects the work functionof the metal surface. This is defined as the work required to remove an electron from themetal to the surface. Due to the lack of nearest neighbour interactions for surface atomscompared to metal atoms in the bulk, electrons in the surface atoms have higher energy andtend to “spill-over” onto the surface leading to a negative charge on the surface (Figure 2.6top). This results in a dipole facing away from the metal, which opposes electron movementto the surface, leading to an increase in the work function. Introducing atomic roughness tothe surface leads to a competing process where, instead of the distribution of net surfacecharge following the rough features on the surface, it “smoothens” out.23 In this process, thenegative charge on sharp features moves from the sharp features into “valleys”, leaving behinda positive charge at the sharp features (Figure 2.6 bottom). This results in a dipole pointinginto the surface and a decrease in the work function. A trend exists where the atomicallyrough surfaces have lower work functions. Among the low index planes, the work functionsfor (111), (100) and (110) are 5.26 eV, 5.22 eV and 5.20 eV respectively, from photoemissionspectroscopy measurements.24 Higher index surface crystallographies (210) and (311), havelower work functions than low index planes as they are atomically rougher (5.16 eV and 4.96 eVrespectively, measured via photoemission spectroscopy).25 The differences in work functionfor various surface crystallographies will be shown later to affect the metal-solution interface.In terms of surface crystallography, metal substrates will be described in this thesis inthree ways: as single crystal electrodes, single crystal bead electrodes and a polycrystallinesubstrates (Figure 2.7). Single crystal electrodes are characterized by a single surface crys-tallography such as (111), (100) or (110). The spherical single crystal bead electrodes are112.1. ElectrochemistryFigure 2.6: Charge distribution at the surface due to electron “spill-over” and “smoothing”as a result of the roughened features. Solid lines indicate the rough surface and the dottedlines indicate the charge distribution. The charge distribution, instead of following the surfacefeatures (top), becomes more linear and “smoother” (bottom). This results in the negativecharge in the “valleys” of the rough surface. Adapted with permissions from Smoluchowski,Phys. Rev. 60 (1941), pp. 661-674.23 Copyright 1941 by the American Physical SocietyFigure 2.7: Schematic depicting three types of metal substrates used: (a) Single crystal elec-trode, (b) single crystal bead electrodes and (c) polycrystalline substrates. Length scales areestimates to provide relative sizes of the different substrates.122.1. Electrochemistryprepared from a single crystallization site that grow to produce a spherical surface with dif-ferent surface crystallographies in a well-defined pattern. Frequently used are polycrystallinesubstrates that contain many regions of different surface crystallography separated by grainboundaries. These polycrystalline substrates are made typically either by polishing a goldsurface or sputtering gold onto glass. Studying the influence of surface crystallography us-ing polycrystalline substrates is difficult due to the unpredictable crystallographic compositionof the surface. Electron back scattering diffraction images of polycrystalline gold surfaceshave revealed that these polycrystalline surfaces consist of micron sized grains of low indexsurface crystallographies bordered by grain boundaries (Figure 2.8).26 The specific surfacecrystallography of the individual grains is difficult to control. However, annealing is typicallyused to maximize the size of the grains and reduce the variety of surface crystallographic fea-tures. As lower energy surfaces are preferred, low index planes (e.g. Au(111), Au(100) andAu(110)) would predominate on annealed surfaces.27,28 Grain boundaries are difficult to indexbut are considered as behaving like higher index surfaces such as (311) and (210). Utilisingsingle crystal substrates allows analytical studies of the influence of surface crystallography.The variety and distribution of surface crystallographies on a polycrystalline material is oftennot considered when making DNA biosensors on commercial metal substrates. Hence, un-derstanding the influence of surface crystallography in manufacturing DNA SAMs will be latershown to be important with preferential adsorption on certain surface crystallographic features.2.1.2 The Metal-Solution Interface and the Electric Double LayerA metal electrode in contact with an electrolyte defines a metal-solution interface that is per-tinent to electrochemical measurements. When an electric potential is applied to a metalsurface (φo) (with respect to a solution potential), two processes can occur: (1) A non-faradaicprocess where the metal surface becomes charged results in the ions in solution reorganizingto compensate the charges. (2) A faradaic process occurs that is the reduction or oxidationof redox active species present at the surface. This results in an electron transfer across themetal-solution interface. To begin, non-faradaic processes will be considered first for a metalin an electrolyte containing a 1:1 ratio of cations to anions.29132.1. ElectrochemistryFigure 2.8: a) Lateral force microscopy measurements of a gold polycrystalline surface (pre-pared from a polycrystalline gold bead by polishing then annealing) with the (b) electron backscattering diffraction image of gold surface shown. (c) Colours in image b correspond to theindexed surface crystallography. Quadrant I primarily consists of Au(111) surfaces, quadrant IIconsists of both Au(100) and Au(110) surfaces while quadrant III and IV consist of more higherindex surface crystallographies. Adapted from Smith and colleagues.26 Copyright (2012), withpermission from Elsevier.For example at a specific applied potential, the surface of the metal can have a positivecharge (σm). This will be compensated by a layer of anions near the surface with an equivalentcharge of σs. The large size of hydrated ions limits the number of anions that can resideimmediately next to the surface. The lower conductivity of the electrolyte compared to themetal also means that more ions than are next to the surface are needed to compensate thecharge of the metal surface. This results in a concentration of ions near the metal surface thatis greater than in the bulk electrolyte. This ion distribution is modeled by two layers that extendinto the electrolyte and is known as the electric double layer (Figure 2.9).30A model of the electric double layer has been proposed individually by Helmholtz, Guoy andChapman then combined by Stern.30 Dubbed, the Guoy-Chapman Stern model, the electricdouble layer is composed of two regions: the Helmholtz layer (Figure 2.9a) and the diffuselayer (Figure 2.9b). The Helmholtz layer consists of the region between the metal surfaceand the plane of closest approach (x2) for a solvated ion, called the outer Helmholtz plane.This region is modeled as a parallel-plate capacitor with the capacitance (C) defined throughequation 2.2:C =εε0x2(2.2)142.1. ElectrochemistryFigure 2.9: Schematic depicting the electric double layer composed of a positive chargedmetal surface, (a) the Helmholtz layer, (b) the diffuse layer and (c) bulk electrolyte. The outerHelmholtz plane is located at the plane of closest approach x2. The potential profile overdistance from the electrode surface is shown below. Specifically adsorbed ions are not shownin the schematic. Water molecules are only shown in the hydration shells of anions and cations.152.1. ElectrochemistryHere ε is the dielectric constant of the medium, ε0 is the permittivity of free space and x2is the space between the parallel plates i.e. the separation between the metal surface andthe outer Helmholtz plane. In the Helmholtz layer, the potential (φ) decreases linearly as in aparallel-plate capacitor and the electric field is given by equation 2.3:dφdx=σεε0(2.3)The additional anions needed to compensate the charge at the metal surface are locatedin a region between the outer Helmholtz plane (OHP) and the bulk solution, called the diffuselayer (Figure 2.9b). The number of anions decrease with distance from the metal surface.This results in an electric field (dφ/dx) that decreases with distance from the metal surface asσm becomes more shielded. In the diffuse layer, φ decays with increasing distance from theelectrode via equation 2.4 where k is Boltzmann’s constant, T is the temperature, n0 is thebulk anion concentration in number of ions per liter, e is the elementary charge and z is thecharge of the anion.dφdx= −(8kTn0εε0)1/2sinh(zeφ2kT) (2.4)The concentration of ions in the electrolyte has an effect on the electric field and the dis-tance dependent potential profile. In a physical sense, at higher ion concentrations, more ionsare present to shield the charge of the metal surface, leading to the decay of the potential oc-curring at a distance closer to the metal surface. The distance from the surface for the potentialto decay is related to the Debye length (κ−1), with its inverse, κ , shown in equation 2.5. Theeffective thickness of the diffuse layer (99.99% charge compensating σm) can be calculatedthrough ln(104)× κ−1 and is shown to decrease with increasing electrolyte concentration andionic strength (Figure 2.10).κ =(2n0z2e2εε0kT)1/2(2.5)The capacitance in the diffuse layer depends on the potential at the metal surface (φ0) andthe ion concentration (n0) as defined by equation Electrochemistry 0 1000 2000 3000 4000 5000 6000 0.1  1  10  100τ eff (nm)Ion Concentration (mM)z=1z=2z=3Figure 2.10: The calculated effective thickness of the diffuse layer decreases with increasingelectrolyte concentration. This is shown for electrolytes containing monovalent, divalent ortrivalent cations or anions (or z=1,2, or 3). The relation τ99.99% = ln(104) × κ−1 is used toillustrate this point.C =(2z2e2εε0n0kT)1/2cosh(zeφ02kT) (2.6)The total capacitance (CT ) of the Guoy-Chapman-Stern model can be modeled as twocapacitors in series: one for the Helmholtz layer and one for the diffuse layer as in equation2.7.1CT=1CH+1CD=x2εε0+1(2z2e2εε0n0kT)1/2cosh( zeφ02kT )(2.7)The capacitance in the electric double layer is dependent on the dielectric constant, thatis affected by the composition of ions at the interface as well as adsorbed species (e.g. ions,organic molecules). The capacitance also changes as function of potential and concentrationof ions in the electrolyte. If an infinite potential is applied, the number of ions at the surface islimited by the amount of hydrated ions which can exist in the OHP. This results in the capac-itance becoming a maximum at large φ0 or high σm (Figure 2.11a). Based on the model, a172.1. Electrochemistry 0 50 100 150 200-1 -0.5  0  0.5  1C (µF/cm2φ0 (V)1 M0.1 M10 mM1 mM(a) (b)Figure 2.11: (a) The calculated capacitance from the Guoy-Chapman Stern model as a func-tion of potential φ0 for a metal surface in 1 mM to 1 M electrolyte containing monovalent cationsand anions. (b) Differential capacitance measurements of a Hg electrode in 1 mM to 1 M NaFelectrolyte. Reprinted with permission from Allen J. Bard Larry R. Faulkner.33 Copyright (2001)John Wiley and Sons. Adapted with permission from Grahame.31 Copyright (1947) AmericanChemical Society.minimum capacitance is observed when φ0 is zero. This is the potential of zero charge (pzc),where the charge of the metal surface is zero and charge compensation by ions is not needed.The lower amount of ions present results in a lower capacitance value, which decreases withlesser ionic strength. The capacitance minimum can be observed experimentally on real metalsurfaces in the absence of specifically adsorbed ions or molecules (Figure 2.11b).31,32 Thepotential where this capacitance minimum occurs is the potential of zero charge of the metalsurface.The electric double layer model does not take into account the effect of adsorbing species.Self-assembled monolayers, which will be discussed in a later section, are composed of or-ganic molecules adsorbed onto metal surfaces. They are known to shift the pzc and changethe dielectric in the double layer. The effect of specifically adsorbed anions on the electricdouble-layer is also worth discussing empirically as it is pertinent to Chapter 7. Generally, an-182.1. ElectrochemistryFigure 2.12: Potential of zero charge (#) varies across the surface crystallographies along thezone axis (110) to (111) to (100) to (110). A correlation between the density of “broken bonds”(+) and the potential of zero charge is evident. Reprinted from deLevie.35 Copyright (1990)with permission from Elsevier.ions with a strong hydration shell, such as F– , ClO –4 and PF–6 interact with the metal surfacethrough electrostatic forces.34 Anions that have weak hydration shells such as Cl– , Br– and I–form a chemical interaction with the surface. The number of specifically adsorbed anions at themetal surface often exceeds the number of anions normally required to compensate σm. Thisresults in a negative shift of the pzc that depends on the concentration of specifically-adsorbinganions in the bulk electrolyte.In addition to the electrolyte composition, the metal surface morphology can also affect thepzc. The atomic roughness of a metal surface, which is related to the surface crystallogra-phy, influences the work function due to electron “spill-over” effect and electron “smoothing”.23Atomically rougher surface crystallographies and surfaces with a higher density of “brokenbonds” have lower work functions, which is closely correlated with the pzc.25 Surfaces with ahigher work function have a higher potential of zero charge in non-adsorbing electrolyte. Forgold, the (111), which is the most atomically smooth surface, has the most positive pzc andthe (210), which is the most atomically rough surface, has the most negative pzc. As per therelation of work function to atomic roughness, the point of zero charge decreases in order from(111), (100), (311), (110) and (210).35While the Guoy-Chapman Stern model can only illustrate an ideal electric double-layer,electrochemical measurements, such as differential capacitance, have been used to study192.1. Electrochemistryand then describe real metal-solution interfaces. Capacitance along with other electrochem-ical measurements will be explained in a later section with details on how they are used tocharacterize adsorbed ionic and organic species at the surface. Up to now, only non-faradaicprocesses have been discussed in describing the electric double layer model. Faradaic pro-cesses, which involve electron transfer from redox-active species and the diffusion of ions atthe metal-solution interface will be discussed in the next section.2.1.3 Faradaic ProcessesIn this thesis, redox-active species are used to label and characterize DNA SAMs throughmeasurement of faradaic currents. A redox active specie is a molecule or ion that can undergoan electron transfer (Equation 2.8).Oxn + e−  Redn−1 (2.8)The current during this redox process depends on the rate of reduction and oxidation (surface concentration of Redn−1 and Oxn) shown in equation 2.9. Here, n is the numberof electrons being transferred in the reaction, F is Faraday’s constant, A is the area of theelectrode surface and kox and kred are the rate constants for oxidation and reduction .i = nFA(kox[Redn−1]x=0 − kred[Oxn]x=0) (2.9)The rate constants for oxidation and reduction are influenced by the overpotential (η), whichis the difference between the applied potential (E) and the potential where the net current iszero (Eequil). The rate constants are defined by equations 2.10 and 2.11. k0 is the rate con-stant of oxidation and reduction at equilibrium when E = Eequil, α is called the charge transfercoefficient and indicates the symmetry of the energy barrier.36 The overall current is then re-lated to the overpotential (η = E − Eequil) as shown in equation 2.12. Applying potentialsbelow Eequil increases the rate of reduction, generating a negative cathodic current and po-tentials above Eequil increases the rate of oxidation, generating a positive anodic current. Theredox active species that pertain to this thesis have fast redox kinetics and are reversible. In202.1. Electrochemistryother words, k0 is large and therefore, the magnitude of the overpotential does not need to belarge in order to generate a net current.kred = k0 exp((−α)F (E − Eequil)RT) (2.10)kox = k0 exp((1− α)F (E − Eequil)RT) (2.11)i = nFAk0([Redn−1]exp(1− α)FηRT− [Oxn] exp −αFηRT) (2.12)At large overpotentials, the rate of electron transfer becomes limited by the diffusion ofredox-active species to the surface. This results in a maximum current that is defined byequation 2.13. This mass-transport limited current depends on the concentration gradient ofthe redox active species and its diffusion coefficient D (Equation 2.13).i = nFAD(∂C(x, t)∂x)x=0 (2.13)With large magnitude overpotentials, all reactants at the surface are depleted resulting in aconcentration gradient. Over time, this concentration gradient becomes lower with the squareroot of time. As a result, the current also has a dependence on the square root of time asshown in equation 2.14. This particular time-dependence will be later used to differentiatebetween redox species diffusing to the surface and redox species adsorbed to the surface.i(t) =nFAD1/20 Cbulk(pit)1/2(2.14)Measuring these faradaic currents have been used to characterize the environment at themetal-solution interface. Redox-active species in solution can be used to characterize theblocking capabilities of adsorbed material on the metal surface. Adsorbates on the surfacecan be labelled with redox active moieties to characterize their properties. The next section willexplore different types of electrochemical measurements where a potential profile is appliedand both faradaic and non-faradaic currents are measured.212.1. Electrochemistry2.1.4 Electrochemical MeasurementsElectrochemical measurements are a common approach for characterizing metal surfaces orexamining systems containing redox active species. Typically these involve applying a potentialto an electrode and measuring the current that flows through the metal-solution interface. Inan electrochemical cell, the electrode being studied is called the working electrode that isplaced in an electrolyte with a reference electrode (Figure 2.13a). This 2-electrode set-up canbe modeled as a (RC)R circuit (Figure 2.13b), where non-faradaic processes are similar tothe charging of a capacitor with capacitance Cdl, faradaic processes are modeled by currentpassing through a resistor with resistance Rct that is in parallel to the capacitor. The referenceelectrode and the working electrode are separated by a resistance (Rs) that is related to theconductivity of the electrolyte.2-electrode systems are generally used only when potentials are applied and low currentsare measured. If large currents are generated and Rs is large, then the potential across theelectric double layer is only a portion of the potential applied between the reference and work-ing electrode. This problem is solved by the addition of a counter electrode through which thecurrent passes. Therefore, the potential at the working electrode with respect to the referenceis the same as the potential across the electric double layer (Figure 2.13c).In this work, three electrochemical measurements will be used, with each technique in-volving a different potential profile. Potentiostatic measurements, where the potential is keptconstant, as done in chronocoulometry (Figure 2.14a), will be shown first. This will be fol-lowed by a linear variation of potential as in cyclic voltammetry (Figure 2.14b). Electrochemicalimpedance spectroscopy will be described last (Figure 2.14c). The current response measuredusing these techniques will be explained for both non-faradaic and faradaic processes. ChronocoulometryApplying a constant potential and measuring the resulting current is a chronoamperometricmeasurement. Integrating the measured current over time to obtain a charge results in achronocoulometric (CC) measurement. In the absence of faradaic processes, the electro-chemical cell can be modeled as a RC circuit. The initial maximum current depends on the Rs222.1. ElectrochemistryFigure 2.13: a) Schematic of a typical 3-electrode electrochemical cell with a working electrode(WE), reference electrode (RE) and counter electrode (CE) if necessary. (b) The correspond-ing simplified circuit model (RC)R depicting faradaic and non-faradaic processes. (c) If thesolution resistance Rs is high, then a counter electrode (CE) is added alleviating the resis-tance between the reference and working electrode.Figure 2.14: The potential profile applied for electrochemical measurements: (a) chronoam-perometry (b) cyclic voltammetry and (c) electrochemical impedance spectroscopy. The typi-cal readout for each system is shown below: (a) the measured current over time, (b) a cyclicvoltammogram and (c) a Nyquist plot.232.1. ElectrochemistryFigure 2.15: Example chronocoulometric response where a gold electrode with Ru(NH3)3+6adsorbed to the surface is immersed in an electrolyte containing Ru(NH3)3+6 . Reprinted fromGe and colleagues.37 Copyright (2007) with permission from Elsevier.through Ohm’s law. Over time, the measured current decays to zero as the capacitor becomescharged (to Qdl). The current profile i (Equation 2.15) depends on the product of Rs and Cdl,also known as the time constant (τ ).i(t) =ERsexp(−tRsCdl) (2.15)Redox active species adsorbed to the electrode surface changes the current response.Provided that the overpotential is large enough, the redox active species are all immediatelyreduced or oxidized. The current and the charge generated are proportional to the number ofredox active species present per unit area (ΓO). Reduction or oxidation of redox-active speciesdiffusing from solution will generate a mass-transport limited current. From equation 2.14, thecurrent is proportional to the inverse of the square root of time and the charge increaseslinearly with the square root of time.Under these conditions, the total charge (Q) is composed of the electron transfer reactionof the redox active species diffusing from solution to the surface, charging of the double layer(Qdl), and electron transfer from the redox-active species adsorbed on the surface (Equation2.16).242.1. ElectrochemistryQ =2nFACbulko D1/2pi1/2t1/2 +Qdl + nFAΓO (2.16)When time is zero, no mass-transport charge is present. The total charge is made dueto the charging of the double layer and the reaction of redox-active species at the surface.The charge of the double layer (Qdl) can be obtained from a chronocoulometric measurementwithout redox-active species present. Subtracting Qdl from the total charge at zero time allowsquantitation of redox-active species present. In Figure 2.15, the coverage of redox-activespecies on the electrode surface can be found from the difference in y intercept when redox-active species are present and with an identical sample without the presence of redox activespecies. Cyclic VoltammetryIn cyclic voltammetry (CV), the potential is linearly varied at a given scan rate (v) and thecurrent is measured. The output called a cyclic voltammogram where the current is plottedagainst the potential (Figure 2.16). In CV measurements without redox-active species present,linearly increasing the potential charges the double layer generates a positive current. Thepotential increases until it reaches a positive limit then the direction is reversed. The potentiallinearly decreases, resulting in a negative current that is equal in magnitude to the positivecurrent. The potential changes until a negative limit is reached and the potential is linearlyincreased again. Thus the potential is cycled between the positive and negative limits.With redox active species adsorbed to the surface, the current depends on redox kineticsand is proportional to the overpotential, the scan rate (v) and the coverage of adsorbed redox-active species (ΓO) as in equation 2.17.38i =n2F 24RTvAΓOexp[ nFRT (E − Eequil)]{1 + exp[ nFRT (E − Eequil)]}2(2.17)The cyclic voltammogram measured is shown in Figure 2.16 where the cathodic current andanodic currents have peaks centered at Eequil. The cathodic peak and anodic peak are alignedso long as the electron transfer is fast and electrochemically reversible. The peak current (ip)252.1. Electrochemistryfor both the anodic and cathodic processes are proportional to the coverage of redox activespecies at the surface (ΓO) and the scan rate (v) as shown in equation 2.18.ip =n2F 24RTvAΓO (2.18)It should be noted that the potential where the current peak Ep occurs with respect to Eequilchanges depending on the relative strength of adsorption of the redox active species to theelectrode surface. If both the Redn−1 and Oxn adsorb to the surface to the same extent, thenEp = Eequil. If Oxn adsorbs more strongly, then Ep < Eequil and if Redn−1 adsorb morestrongly, then Ep > Eo. Furthermore, the peak shape shown in Figure 2.16a is observed forideal redox-active species, but the peak width can be narrowed or broadened by intermolecularinteractions.39For redox active species in solution, the electron transfer current is dependent on bothredox kinetics and mass-transport of the redox-active species to the surface. Hence, it isdependent on the overpotential, the concentration of redox active species in the bulk solutionand the scan rate (v). In a CV measurement, two peaks are seen corresponding to oxidationand reduction centered around Eequil. A peak is observed due to the current decreasing as itbecomes mass-transport limited (Figure 2.16b). The peak current is proportional to the squareroot of the scan rate as shown in equation 2.19 .ip = (2.69× 105)n3/2AD1/2C∗ov1/2 (2.19)This is diagnostic of the redox-active species diffusing to the electrode surface. The peakpotentials (Ep) are unaffected by the scan rate for a reversible electron transfer. The potentialof the anodic peak is approximately 57 mV higher than potential of the cathodic peak for a oneelectron redox centered around Eequil.40From a CV, the area of the current peak divided by the scan rate is the charge passedduring the redox process. The charge is proportional to the amount of redox active speciesbeing reduced or oxidized. The peak current dependence on the scan rate can be used to dis-tinguish adsorbed redox-active species from redox-active species from solution. The coverageof redox active species on the surface (such as Ru(NH3)3+6 ) can be determined by measuring262.1. ElectrochemistryFigure 2.16: Example cyclic voltammogram of reduction and oxidation ofRu(NH3)3+6 /Ru(NH3)2+6 (a) adsorbed at the electrode surface and (b) diffusing from so-lution. Reprinted from Ge and colleagues.37 Copyright (2007) with permission from Elsevier.the charge of the appropriate peak. Electrochemical Impedance SpectroscopyElectrochemical impedance spectroscopy or EIS is another voltammetric technique used tostudy electrochemical systems. EIS measurements involve the application of an AC sine wavepotential with a given amplitude and frequency (f ) (Figure 2.14c).For an electrochemical system with no faradaic processes, a series RC circuit is used as amodel. The capacitance (C) is inversely proportional to the potential as in equation 2.20 andthe resistance (R) is directly proportional to the potential as in Ohm’s law (Equation 2.21).C =σE(2.20)E = iR (2.21)Applying an AC potential with a given angular frequency (ω = 2pif ) as in equation 2.22across either a resistor or a capacitor results in different current responses. The currentthrough the resistor (Equation 2.23) is in-phase with the applied potential, while the currentthrough the capacitor (Equation 2.24) is out-of-phase, lagging by pi/2 radians from the applied272.1. Electrochemistrypotential.E(t) = EAC sin(ωt) (2.22)i(t) =EACRsin(ωt) (2.23)i(t) = CdE(t)dt= CωEAC cos(ωt) = CωEAC sin(ωt+pi2) (2.24)When applying an AC sine wave potential to a series RC circuit, the potential across theresistor and the capacitor add together and is related to the current through a vector called theimpedance (Z).E = ER + EC = iZ (2.25)The impedance has contributions from real (Zreal, in-phase) and imaginary (Zimg, out-of-phase) components (Equation 2.26).Z = Zreal − jZimg = R− jωC(2.26)The impedance depends on the frequency of the potential perturbation. Z can be drawn on avector diagram which has Zimg perpendicular to Zreal as shown in Figure 2.18. The impedanceZ is a vector with length |Z| oriented at an angle φ (phase shift) with respect to Zreal.An EIS measurement consists of measuring Z at many frequencies (f ) and the resultingoutput for a series RC circuit is shown in Figure 2.18. Plotting Zimg vs. Zreal results in aNyquist plot. Plotting the log of the magnitude of Z as a function of log frequency results in aBode plot. A Bode phase diagram is the change in φ as a function of log frequency. TypicallyEIS data is fitted to simulated data from a circuit model to obtain a capacitance and resistancevalue.282.1. ElectrochemistryFigure 2.17: Vector diagram indicating the impedance vector Z. 0 50 100 150 200 0  50  100  150  200Z img(Ω)Zreal(Ω) 100 1000 10000 100000 1  1000 1x106|Z|(Ω)Frequency (Hz) 0 10 20 30 40 50 60 70 80 90 1  1000 1x106-Phase (o)Frequency (Hz)Figure 2.18: Simulated EIS data from a series RC circuit with a 100 Ω resistor and 1 µFcapacitor. The Nyquist plot (left), Bode plot (middle) and Bode phase diagram (c) are outputsfrom an EIS measurement.292.2. Self-Assembled MonolayersTo characterize the metal-solution interface, differential capacitance measurements aretypically performed. These are essentially EIS measurements done at a single frequencywhile scanning the potential. The resulting in-phase and out-of-phase currents (measurablewith a lock-in amplifier) can be used to calculate a double layer capacitance and the solutionresistance assuming a series RC circuit model. Differential capacitance measurements canbe made provided no faradaic processes are present and at a low enough frequency that theresulting impedance has a phase close to 90 degrees.To characterize faradaic processes using EIS, the potential perturbation must center onthe open circuit potential. To interpret the EIS response, a (RC)R circuit model is used. Athigh frequencies, the impedance across the capacitor is effectively zero, resulting in the to-tal impedance being dominated by Rs. As the frequency decreases, the impedance from thecapacitance ( jωC ) increases creating a semi-circle in the Nyquist plot. At low frequencies, aninfinite impedance across the capacitor results in current flow only through Rct and Rs. Anexample Nyquist plot is shown in Figure 2.19 for a circuit model of an electrochemical sys-tem involving redox-active species. Model parameters describing the electrochemical cell areevident in the Nyquist plot. The first x intercept being Rs and the 2nd x intercept at lowerfrequencies corresponding to Rs + Rct. In a real electrochemical cell mass-transport limitedcurrents dominate at low frequencies resulting in a contribution from what is called the Warbergimpedance. This results in the same increase in the real and imaginary impedance. Overall,the metal-solution interface and the kinetics of redox processes can be simultaneously char-acterized with EIS measurements.2.2 Self-Assembled Monolayers2.2.1 Formation of Self-Assembled MonolayersSelf-assembled monolayers (SAMs) consist of organic molecules that adsorb onto metal sur-faces. The adsorption is spontaneous due to lowered metal surface energy when the moleculesadsorb. SAMs form by exposing a clean metal substrate to adsorbates from vapour or solution.The simplicity and lack specialized of equipment needed makes self-assembly a favourable302.2. Self-Assembled Monolayers 0 100 200 300 400 500 600 100  200  300  400  500  600  700Z img(Ω)Zreal(Ω)Figure 2.19: EIS data simulated from an electrochemical system modeled as a R(C[RW])circuit (a.k.a. Randle’s circuit).method for modifying metal surfaces.8,11The adsorbates in SAMs are composed of three components: a headgroup that binds tothe surface, a spacer containing an alkane and a terminal group which influences the surfaceproperties (Figure 2.20). In this thesis, only alkanethiol self-assembled monolayers on goldare used and discussed. The adsorption of alkanethiols on gold occurs via an oxidation reac-tion shown in equation 2.27. The process occurs as follows: the alkanethiol non-specificallyadsorbs onto the gold surface which subsequently leads to partial oxidation of the gold surface(80% electron transfer based on measurements during the adsorption of an octadecanethiolSAM).41 The fate of the hydrogen is the subject of debate and is often thought to either bepresent on the surface or departs from the surface as a proton or hydrogen gas.8RSH + Ausurf −−→ RSAusurf + H+ + e− (2.27)An alkanethiol SAM can also be made by adsorbing disulphides via equation 2.28. Thoughthe resulting SAM is the same in composition to the alkanethiol SAM made using equation2.27, the adsorption kinetics has been observed to be more complex due to the reduction ofthe disulphide bond.42312.2. Self-Assembled MonolayersFigure 2.20: Schematic of a alkanethiol SAM containing mercaptohexanol on a gold surface.RSSR + Ausurf −−→ 2 RSAusurf (2.28)The surface coverage (Γ) is a way of characterizing the extent of adsorption and has unitsthat are in mol/cm2. Two factors contribute to the stability of alkanethiol self-assembled mono-layers on gold: the Au thiolate bond and the van der Waal forces between the alkane chains.The hydrocarbon chains must be near each other in order for the van der Waal forces to beeffective, hence low coverage SAMs are more susceptible to degradation. Furthermore, SAMscontaining long chained alkyl adsorbates have been found to be more stable than SAMs com-posed of shorter alkyl chains due to the greater number of van der Waal interactions.8 Thefunctional group at the terminal end of the alkane-thiol absorbates can contribute to the SAMstability with carboxylic acids and hydroxyl groups self-interacting through hydrogen bonding.43The Au-thiolate bond, which has a bond energy of ~50 kcal or ~200 kJ/mol, is considereda covalent interaction (>100 kJ/mol).8 Oxidation of the thiolate into sulfoxides over time orcatalyzed by UV light leads to the degradation of these alkanethiol SAMs.8,44,45 Desorption asa result of oxidation has been seen at defect regions (step edges, domain boundaries, andsimilar discontinuities and defects).45 Minimizing these defect regions is key to a stable SAM.Without UV oxidation, SAMs are stable from several hours to several days.The alkanethiols quickly adsorb onto the surface at high coverages within minutes, afterwhich reorganization of the layer can occur from 12-24 hours. The initial step in SAM formationis described to be a diffusion controlled adsorption depending on the concentration of thiol in322.2. Self-Assembled Monolayerssolution much like Langmuir adsorption.11 In the Langmuir adsorption isotherm, the rate ofadsorption is dependent on the fractional number of unoccupied sites on the surface (1−θ) andadsorbate concentration (CA) in solution or in vapour while the rate of desorption is dependenton the fractional coverage (θ). Equation 2.29 depicts the overall relation between the fractionalcoverage and the concentration of adsorbate.θ =KeqCA1 +KeqCA(2.29)During the assembly of SAMs from vapour, it has been observed, by scanning tunnelingmicroscopy at low coverage, that the alkane chain lies on the surface parallel to one another inwhat is called the striped phase.46 Prior to the striped phase, it is speculated that the moleculesare physisorbed, in other words, not adsorbed via the Au thiolate bond. This is followed bynucleation and growth where subsequent alkanethiols adsorb nearby an already adsorbedalkanethiol eventually forming island domains. This changes to the alkanethiols becomingupright as the coverage increases. This phase transition is the rate determining step especiallyfor longer alkanethiol chains. With an annealing process where the alkane chains rotate andgold thiolates are mobile on the surface, the alkanethiol SAM becomes more crystalline withmaximized van der Waal interactions.11,47–49Forming SAMs from solution is similar to deposition from vapour, where a lying down tostanding up transition occurs. The striped intermediate phase where the alkanethiols lie onthe surface parallel to one another has not been observed for solution made SAMs. The tran-sition step from the lying down phase to the standing up phase has also been observed withsolution made SAMs and is more prominent with longer alkythiol chains.50 Through reflec-tion absorption infrared spectroscopy and scanning tunneling microscopy, the SAM appearedto grow as small islands followed by domain formations that grew with immersion time.51,52Annealing and reorganization over time is also evident in SAMs made from solution.53Forming mixed SAMs, or SAMs composed of more than one molecule, provides furtheropportunities for modifying metal surfaces. Mixed SAMs can be made through two processes:codeposition, where two adsorbates exist in the initial deposition solution or via thiol exchange,where an initial SAM is formed then is partially exchanged by a different alkanethiol.Thiol-332.2. Self-Assembled MonolayersFigure 2.21: Spatial arrangement of an alkanethiol SAM on Au(111) with (√3 × √3)R30◦organized in c(4 × 2) lattices (red) or 2√3 × 3 lattices (blue). Reprinted with permission fromLove and colleagues.8 Copyright (2005) American Chemical Society.exchange has been characterized to be a pseudo first order reaction dependent on the con-centration of secondary thiol. Thiol-exchange occurs in two steps: a fast replacement at do-main boundaries and defect sites, followed by slower replacement of higher packed alkanethi-ols.54,55 The replacement of a shorter alkane thiol with a longer alkane thiol is faster comparedto the reverse, with the stability of layers and compactness of the layer also affecting the kinet-ics of the exchange.56,57Though the formation of SAMs is affected by the growth process, the structure of SAMshas been found to be affected by a number of factors. The next section will examine thearrangement of the alkanethiol SAMs over various surface crystallographies.2.2.2 The Structure of Self Assembled MonolayersFor stability, the adsorbed alkanethiols must be oriented such that the van der Waal interactionsbetween the hydrocarbon backbones are maximized. Through modeling and confirmation withreflection absorption infrared spectroscopy measurements, alkanethiols in SAMs were foundto adapt a trans configuration, with the alkanethiols tilting towards their nearest neighbour.8In regards to the organization of the alkanethiols on the gold surface, early evidence fromdiffraction and STM measurements of alkanethiol SAMs on Au(111) suggested that the alka-nethiol arranges on the gold with a (√3×√3)R30◦ overlayer.58,59 The tilt angle for these alka-342.2. Self-Assembled MonolayersFigure 2.22: Schematic showing the Au-adatom-dithiolate structures which exist in methylthi-olate SAMs. Yellow atoms indicate gold surface atoms, red atoms indicate Au adatoms, blueatoms indicate sulphur, and black indicate carbon atoms. Adapted with permission from Tangand colleagues.60 Copyright 2012 with permission from Elsevier.nethiols were measured to be 270 from the normal of the Au(111) surface. This overlayerarrangement can also be described as a hexagonal close-packed formation. Density func-tional theory calculations surmised that the sulphur atom resides in the hollow-site betweenthe 3 gold atoms(Figure 2.21).8 A c(4×2) lattice or a 2√3× 3 lattice was also used to describethe organization of the SAM when incorporating alkanethiols chains into the symmetry.In light of more recent experimental evidence, other modes of sulphur binding with the goldsuch as bridge sites (where the sulphur is located between two atoms) have also been pro-posed.61 STM measurements of methylthiolate SAMs indicated (3× 4) structures containingAu-adatom-dithiolate structures (Figure 2.22).60,62 The gold adatom is present as a result ofthe reconstruction of the gold surface. As previously mentioned, reconstruction of the Au sur-face can occur where surface atoms rearrange to lower the overall surface energy. The changein the surface atomic arrangement is different from that which was dictated by the underlyingsurface crystallography. Upon adsorption of the alkanethiol on Au(111), the lifting (i.e. rever-sal) of the surface reconstruction results in atoms being pulled above the gold surface creatingadatoms. As a result, gold islands and missing gold atoms (a.k.a. vacancies) in the surfacehave been detected underneath the alkanethiol SAMs.22 For SAMs made with alkanethiolscontaining longer carbon chains than propanethiols, STM measurements have detected do-mains of (√3×√3)R30◦ and c(4× 2) lattices with the former containing vacancies in the goldsurface and both containing Au-adatom-dithiolate structures (Figure 2.23).61Alkanethiol SAMs has also been observed to undergo reorganization with (6 × √3) super-lattices 3 months after SAM formation.63 Undoubtedly, surfaces of different crystallographic352.2. Self-Assembled MonolayersFigure 2.23: Schematic showing the lattices (√3×√3)R30◦ and c(4× 2) that have been de-tected via STM for long chain alkanethiol containing SAMs. Diagram depicts the most thermo-dynamically stable structure based on modeling of a methylthiolate SAM. The sulphur (greenatoms ) adsorbs onto the gold surface (yellow atoms) with a given number of gold adatoms(white atoms), and vacancies in the gold surface (red atoms). Adapted with permission fromPensa and colleagues.61 Copyright 2012 with permission from American Chemical Society.character will change the arrangement of the alkanethiols on the surface as well as the tiltangle of the alkanethiols.Since the Au-thiol bonding arrangement will be different on other surface crystallographies,alkanethiol SAMs on Au(100) have also been studied, although to a lesser extent compared toSAMs on Au(111). With molecular dynamic simulations, the alkanethiol SAMs were rational-ized to have closer packing on Au(100) compared to Au(111).64 Alkanethiols were thought toadsorb onto Au(100) in square c(2×2) lattices with a tilt angle 140 from the normal of the goldsurface, which is smaller compared to the tilt angle of alkanethiols on Au(111) (270).8 Evidencefrom He diffraction58 and STM measurements65 indicate that alkanethiols SAMs on Au(100)are packed at higher coverages than the Au(111) via an oblique array that did not align withthe underlying surface crystallography. However, the c(2×2) lattices which were previouslypredicted by simulations were not visible in the STM measurements. Lifting of the surfacereconstruction of Au(100) has also been observed when adsorbing alkanethiols. Evidencesuggest that the reconstruction is not completely reversed, leaving domains of the alkanethiolSAM to be organized in a hexagonal pattern on Au(100).65Au(110) undergoes reconstruction where the surface contains a “missing row” of atoms.Diffraction measurements of alkanethiol SAMs on Au(110) indicate lifting of the reconstructionand the alkanethiol organizing into square c(2×2) lattices on Au(110) surfaces with the tilt angleof the alkanethiols being ~370.58 Not much experimental evidence is available for alkanethiol362.2. Self-Assembled MonolayersSAMs on other surface crystallographies especially higher index surfaces. This is due to theinconvenience in forming single crystal substrates of higher index and the lack of motivation forstudying these specialized substrates. A lower amount of high index surfaces are present ongold on glass substrates. Hence, the likelihood of forming an alkanethiol SAM on these higherindex surfaces is of lesser interest compared to SAMs on lower index surfaces. However, aswill later be seen, indirect methods can help in understanding the SAM structures on thesehigher index surface crystallographies.Scanning probe microscopy studies have detected defects in alkanethiol SAMs assem-bled onto a given surface crystallography (e.g. Au(111)) and prominently at domain bound-aries.66 Though these defects for the most part are unavoidable, they are less common inSAMs assembled from vapour, possibly attributing the cause of some defects to various sol-vents. Defects in the SAM include missing rows, pinholes, collapsed sites and molecularvacancies.22,66–68 Often defects in the SAM result in the alkanethiols not adopting an all-transconfiguration and therefore having decreased van der Waal interactions.Knowledge of the structure and arrangement of the alkanethiol SAM is highly valuable.However, characterizing SAMs using readily available instrumentals is required in order toutilise alkanethiol SAMs for other applications.2.2.3 Electrochemical Measurements of SAMsCharacterizing SAMs and monitoring their growth has been accomplished with surface plas-mon resonance, infrared spectroscopy, scanning tunneling microscopy among other meth-ods. With electrochemical characterization of SAMs being featured prominently in this thesis,how various electrochemical responses are affected by the presence of an alkanethiol self-assembled monolayer will be discussed.Alkanethiol SAMs, which reside between the OHP and the metal surface, will affect theelectric double layer. The organic alkanethiol is less conductive than the aqueous electrolyte,resulting in a lower dielectric constant at the interface and a smaller measured capacitance.Since the capacitance in the Helmholtz layer will be smaller, the electric field (dφ/dx) will in-crease based on equation 2.3 and the potential at the outer Helmholtz is much closer to the372.2. Self-Assembled Monolayerspotential in solution. With a lower dielectric constant, the net dipole moment at the surfacechanges. As previously mentioned, the surface dipole determines the work function, which isclosely related with the pzc. The presence of an alkanethiol SAM shifts the pzc of the goldsurface to more negative potentials. Alkanethiol SAMs composed of longer carbon chains canshift the pzc as much as -0.5 V compared to bare gold.69 SAMs with hydroxy terminated alka-nethiols have different surface dipoles, resulting in a slight positive shift in the pzc comparedto bare gold.69The smaller capacitance, indicating the presence of the SAM on the surface, can be usedto measure the fractional coverage (θ) through equation 2.30. The Frumkin parallel platecapacitor model is used to describe the partially covered SAM, with the surface made upof multiple capacitors in parallel. Each of these capacitors have a capacitance that correspondto a 100% coverage SAM (CSAM ) or a capacitance corresponding to bare Au (CAu ).70 Thetotal coverage is a weighted average of all the capacitors, which depends on the coverage ofthe SAM.C = CSAM (θ) + CAu(1− θ) (2.30)An ideal non-conducting SAM prevents electron transfers and therefore faradaic processes.Depending on the nature of the redox active species, redox may be limited leading to smallercurrents in cyclic voltammograms and chronocoulometry. For EIS measurements, the mea-sured Rct becomes larger as electron transfer is hindered.The stability of the SAMs can be affected during electrochemical measurements at extremepotentials where reductive desorption or oxidation of the thiolate occurs. Reductive desorptionof the gold thiol bond is shown in equation 2.31.RSAu + e− −−→ RS− + Au (2.31)The potential which this reductive desorption occurs depends on the number of carbons inthe alkanethiol and the underyling surface crystallography on the gold surface. A larger numberof carbons in the alkane thiol chain results in more van der Waal interactions between thealkanethiols. The increased stability is reflected in the reductive desorption potential becoming382.2. Self-Assembled MonolayersFigure 2.24: CVs showing the reductive desorption of decanethiol SAMs in 0.1 M KOH on(a) polycrystalline gold and (b) monocrystalline surfaces. Alkanethiol SAMs on polycrystallinegold have multiple desorption potentials, indicated by the number of cathodic peaks. Au(111)desorb at potentials less negative than SAMs on Au(210). Reprinted from Doneux and col-leagues.71 Copyright (2010) with permission from Elseviermore negative with the increased number of carbons in the alkanethiol.Properties of the metal surface can also affect the reductive desorption potential. As dis-cussed in the section 2.1.2, atomically flat surfaces have a higher pzc compared to rougherhigh energy surfaces (Figure 2.24).71 Therefore a less negative potential is required to reduc-tively desorb a SAM on Au(111) compared to higher index planes such as Au(210).At increasingly positive potentials, SAMs become unstable due to oxidative desorption.This oxidative desorption results in the sulphur becoming oxidized into sulfoxides with pro-posed pathways shown in equations 2.32 and 2.33. The potential which this occurs rangesfrom 0.6 V to 1.4 V vs. SCE depending on the pH of the solution.44RSAu + 2 H2O −−→ Au + RSO2H + 3 e− + 3 H+ (2.32)392.3. FluorescenceRSAu + 2 H2O −−→ Au + RSO −2 + 3 e− + 4 H+ (2.33)The potential of oxidative desorption in pH 10 on polycrystalline surfaces did not appear tochange depending on surface crystallography72 unlike observed for reductive desorption.73Electrochemistry provides the means to characterize alkanethiol SAMs. However, morepowerful methods exist to image and examine the defects that occur heterogeneously in SAMssuch as scanning probe microscopy methods and optical microscopy techniques.2.3 Fluorescence2.3.1 Fluorescence PhenomenaFluorescence is a spontaneous emission process where an incident photon is absorbed by amolecule, followed by emission of a photon with longer wavelength from the same molecule.74Absorption of the incident photon excites an electron from ground state to an excited state.Prior to emission, non-radiative processes occur where the electron relaxes to a lower excitedstate through internal conversion and/or vibrational relaxation. From this lower excited state,the electron returns to ground state while emitting a photon of lower energy and longer wave-length compared to the incident photon. The likelihood of these electronic transitions dependson the Franck Condon principle, which states electrons can transition between states wherethere is overlap of the wave function. The fluorescence intensity depends on the concentrationof fluorophores present and the quantum yield (QY). The ratio of radiative events over the to-tal rate of both non-radiative events (knr) and radiative events (kr) defines the quantum yield,shown in equation 2.34:QY =krkr + knr(2.34)2.3.2 Quenching at Metal SurfacesLabeling molecules adsorbed to a metal surface with fluorophores provides a means to char-acterize molecules in SAMs. However, fluorescence from an excited fluorophore near a metal-402.3. Fluorescencesolution interface becomes affected due to reflection and adsorption of energy by the metalsurface. To explain this phenomena, one can consider a fluorophore as an oscillating dipole.The oscillating dipole, has a oscillating electric field which emanates from itself.75 Dependingon the distance of the dipole to the surface, an image dipole pointing in the opposite directionappears in the surface. The interaction between the fluorophore dipole and the surface dipoleresults in the movement of electrons in the metal surface which has varying effects dependingon the distance of the fluorophore from the surface.As self-assembled monolayers are typically 1 to 10 nm thick, fluorescence characterizationof SAMs involves measuring fluorophores at these distances from the metal surface. In thisregion, where the distance to the fluorophore is less than 25% of the incident light wavelength,a drop in the quantum yield occurs. This is observed as the fluorophore being quenching bythe metal surface, which has been modeled and rationalized by Chance, Prock and Silbey76and Yeung and Gustafon.77 To summarize their model, at these smaller distances, near-fieldradiation from the oscillating dipole couples with the surface plasmon polaritons in the metal.The result is a non-radiative energy transfer from the dipole to the surface which increaseswith decreasing distance between the fluorophore and the metal surface.The apparent quantum yield for the fluorophore AlexaFluor488 at distances less than 50nm from a metal surface was previously modeled and is shown in Figure 2.25.16 Here, thequantum yield proportionally increases with d3 (Figure 2.25a). This is due to the strength of thedipole coupling with the surface increasing with d−3 resulting in more quenching. To rationalizethe order of this decay, the energy transfer efficiency in Förster resonance energy transfer(FRET) is considered. Here a donor fluorophore which is considered a point undergoes anenergy transfer with an acceptor fluorophore at another point in solution. The FRET efficiencyincreases with d−6 as separation decreases between the acceptor and donor fluorophores.Increasing the dimension of acceptors to a plane array of acceptors results in energy transferdependence on d−4. Acceptors in layers below this plane adds another dimension resulting inan energy transfer dependence that increases with d−3. In the case of a fluorophore near abulk metal surface, this is akin to a 3 dimensional array of acceptors near a donor fluorophore;thus there is a d3 dependence on the quantum yield.75The metal-induced quenching of fluorophores within 50 nm from the metal surface is useful412.3. FluorescenceFigure 2.25: Simulated quantum yield of AlexaFluor488 depending on the distance from ametal surface at distances up to 50 nm. Adapted from Casanova and colleagues.16 Copyright(2017) with permission from Springer Nature.for fluorescence characterization of SAMs. A later section will describe in-situ electrochemicalfluorescence microscopy which heavily relies on metal-induced quenching. This technique hasbeen used to investigate fluorescently labeled alkanethiol SAMs and will be used as a majorcharacterization method in this thesis.42Chapter 3Literature ReviewThis chapter consists of a description of the literature directly pertaining to this thesis. It iscomposed of two parts: DNA self-assembled monolayers, their application to DNA biosensorsand previous attempts to potential-assisted DNA self-assembly; followed by a summary of thehistory and development of in-situ electrochemical fluorescence microscopy image character-ization.3.1 DNA Self-Assembled Monolayers3.1.1 Deoxyribonucleic acidDeoxyribonucleic acid (DNA) is a biomolecule that functions as genetic coding for all livingorganisms. It is composed of two single stranded DNA (ssDNA) bound together to form thedouble helix structure known as double stranded DNA (dsDNA). The structure of DNA waspredicted by Watson and Crick based on X-ray crystallography data collected by Franklin andGosling.78–80ssDNA is made up of repeating units called nucleotides; each contains a sugar (deoxyri-bose), with a phosphate group attached to its 5’ carbon, a nitrogenous base attached to the 1’carbon and the phosphate of the next nucleotide at the 3’ carbon (Figure 3.1). 5’ and 3’ areused to label the beginning and end of the ssDNA. The nitrogenous bases are one of the four:Adenine (A), Thymine (T), Cytosine (C) and Guanine. Adenine and Thymine form a base-pairthrough two hydrogen bonds while Cytosine and Guanine form another base-pair with threehydrogen bonds. The order that these nucleotides (which contain these nitrogenous bases)appear in ssDNA is called a sequence. A ssDNA with a certain sequence interacts with an-other ssDNA containing a complementary sequence, forming base-pairs with between the twossDNA. The two ssDNA interact anti-parallel to one another; the 5’ end of the initial ssDNA433.1. DNA Self-Assembled MonolayersFigure 3.1: Depiction of four nucleotides interacting via hydrogen bonding: cytosine (red) withguanine (green), and thymine (yellow) with adenine (blue). Dashes indicate hydrogen bondingbetween the basepairs. The backbone consists of deoxyribose sugar and phosphate groupswhich alternate along the chain.interacts with the 3’ end of the complement strand and vice versa.Formation of dsDNA from ssDNA is called nucleic acid hybridisation and is not to be con-fused with DNA-DNA hybridisation which is a technique where genetic similarities between dif-ferent species are compared. Nucleic acid hybridisation is an energetically favourable processwhen the ssDNA forms basepairs with the complement strand with the free energy change(∆Go) during hybridisation shown in the following equation.∆Go = −RT ln [dsDNA][ssDNA]2(3.1)For a 30 basepair DNA, the ∆Go of dsDNA formation is close to -35 kcal/mol and at temper-atures between 50 and 60oC, the dsDNA denatures into ssDNA. The temperature at whichthis denaturation occurs is called the melting temperature (TM ). Its formal definition is thetemperature at which 50% of a dsDNA has denatured into two ssDNA and is used to indicatethe stability of a dsDNA. TM increases and ∆Go becomes more negative with the number ofbasepairs in the sequence, with G-C basepairs adding extra stability over A-T basepairs. Amismatch is said to be present when one of the nitrogenous bases in the initial ssDNA cannotform a basepair with its corresponding nitrogenous base in the complement strand. With eachmismatch,the dsDNA structure becomes increasingly less stable as observed through ∆Go be-443.1. DNA Self-Assembled Monolayerscoming more positive and TM decreasing.81 ssDNA can also self-complement through base-pairs within its own sequence to form secondary structures such as hairpin DNA. Since theinteractions are intramolecular, forming such secondary structures is energetically favourable.Such secondary structures have been used for bio-recognition, as will shown in a later section.Due to the negative charge of the phosphate group in the backbone, DNA has a high neg-ative charge density. In an electrolyte containing monovalent cations, the negatively chargedphosphate group is neutralized by a 1 M “cloud” of cations residing near the DNA, independentof bulk cation concentration.82 The cation shields the negative charge with monovalent cationsreducing the repulsive charge of each phosphate group by 76% and divalent cations such asmagnesium reduce the repulsive charge by 88%.82 Shielding the negative charge throughcounter-ion condensation becomes important for hybridisation as indicated by the increasingTM with increasing ionic strength.83The different mechanical properties of ssDNA and dsDNA are important to note whendesigning DNA sensors. One such property different between ssDNA and dsDNA is the per-sistence length, which is defined as the length of the polymer where the vector tangents arealigned.84,85 For ssDNA, the persistence length is 1–2 nm and for dsDNA it is 50 nm.86,87dsDNA that is shorter than the persistence length behaves like a rigid rod. With the length ofdsDNA being 3.4 nm per 10 basepairs,78 dsDNA which contains 150 basepairs behaves likea rigid rod. ssDNA containing 150 bases will behave like a flexible polymer. Increasing thenumber of bases in both ssDNA and dsDNA, increases the entropy and number of configu-rations possible for the DNA.84 It should be noted that the persistence length of both ssDNAand dsDNA decrease with increasing ionic strength as the negative charge of the phosphatebecomes shielded. In an electrolyte containing divalent cations, the ionic strength results in adecrease in the persistence length of dsDNA to 30 nm.84 Decreasing the ionic strength resultsin an increase in the persistence length, with ssDNA behaving like a rigid rod in 10 mM ionicstrength electrolyte.883.1.2 Making DNA Self-Assembled MonolayersBased on alkanethiol self-assembled monolayers research (from section 2.2), DNA SAMson gold, composed of DNA molecules modified with an alkanethiol moiety, were proposed by453.1. DNA Self-Assembled MonolayersFigure 3.2: A schematic showing hexagonal-packed single stranded DNA (grey circles) over Au(111) for a DNA SAMs made in (a) 0.33M NaCl and in (b) 1.2 M NaCl. The effective diameters(deff , dotted circles) of the DNA in both cases were calculated to be 2.5 nm and 2 nm. It isimportant to note that in this DNA arrangement, between adjacent DNA (La), multiple hollowsites remain unoccupied. Adapted from Li and colleagues.92 Copyright (2011) with permissionfrom Elsevier.Herne and Tarlov as a possible template for nucleic acid sensors.89 The DNA adsorbed to thesurface (a.k.a. probe DNA) are designed to imitate primers. In nature, primers are ribonucleicacid oligomers that act as recognition sequences for DNA in the genetic code. This recognitionsequence enables enzymes to also initiate DNA synthesis in-vivo and in-vitro.90 Typically theDNA oligomers immobilized to the surface are at least ~20-50 bases long which exceeds the18-22 base length of primers and enables hybridisation to a complementary strand in solutionwith a sufficient degree of selectivity. As is done during alkanethiol SAM formation, the goldsurface was immersed in a solution containing thiol-modified DNA solution. However, it wasdetermined that most of the DNA was non-specifically adsorbed; the DNA physisorbed ontothe gold through the nitrogenous base pairs instead of chemisorbed through the Au-thiolatechemical bond. Immersing the DNA SAM with non-specifically adsorbed DNA in a solutioncontaining an alkanethiol passivator, such as mercaptohexanol (MCH), displaced the non-specifically adsorbed DNA leaving only chemisorbed DNA.89 It was found that the adsorbedmercaptohexanol forced the tethered DNA to “stand-up” and interact with the solution insteadof the surface.91The organization of DNA SAMs compared to alkanethiol SAMs is more complex due to463.1. DNA Self-Assembled Monolayersthe repulsion between the negative charges on the DNA phosphate backbone. The van derWaal interactions, which play a crucial role in alkanethiol SAMs, do not exist between thenucleic acid bases. Hence, the DNA do not organize in a crystalline manner at the surface.Through molecular dynamics simulations of adsorbed DNA on a silicon surface, Wong andPettitt predicted DNA-DNA spacing to be 2.2 nm apart in a 0.8 M NaCl electrolyte.93 It wasproposed that on Au(111), the alkanethiol moiety on the DNA adsorbs in 3-fold hollow sites withhexagonal packing, similar to alkanethiol SAMs.94,95 More recently, Li and colleagues modeledthe DNA SAM based on the known structures in alkanethiol SAMs and with the assumptionof the helical ssDNA having a cylindrical framework.92 The distance between adjacent DNAmolecules in a SAM is significantly larger than that of alkanethiols due to the negative chargein the phosphate backbone and the wider radius of DNA.15,92 Figure 3.2a shows their modelof a hexagonal-packed DNA SAM on Au(111) made in 0.33 M NaCl. The effective diameterof the DNA (deff ), which accounts for the electrostatic repulsion, and the minimum distancebetween adjacent DNA (La) in this case were calculated to be 2.5 nm, equivalent to 4 adjacenthollow sites.92 Increasing the ionic strength or introducing divalent cations results in the neg-ative charge being shielded. With the repulsion reduced, the DNA-DNA spacing decreases96and the effective diameter of the DNA becomes smaller (Figure 3.2b). Theoretically, the maxi-mum coverage in a DNA SAM would be 3.8× 1013DNA/cm2 and 1.27× 1014DNA/cm2 basedon a 2 nm diameter for dsDNA78 and 1 nm diameter for ssDNA assuming that 100% neg-ative charge of the phosphate backbone is shielded.97 Molecular dynamic simulations havefound the maximum optimal coverage of a ssDNA in 1 M NaCl to be 1.9× 1013DNA/cm2 and1.1× 1013DNA/cm2 for a dsDNA SAM.94 Directly resolving the organization of close-packedDNA has been difficult to do with scanning probe microscopy, due to the long length of DNAand the hydration layer present.98 To the knowledge of the thesis author, only height mea-surements of DNA SAMs or characterization of low coverage DNA SAMs have been achievedthrough atomic force microscopy.98,99Like alkanethiol SAMs, the coverage of the DNA SAMs is dependent on the concentra-tion of DNA in solution. Georgiadis and colleagues monitored the formation of DNA SAMsusing surface plasmon resonance and observed that non-specific adsorption of DNA is fastcompared to the chemisorption of thiol-modified DNA. The alkanethiol moiety was rationalized473.1. DNA Self-Assembled Monolayersto reduce the non-specific adsorption and also increased the desorption rate during the self-assembly process. With the alkanethiol moiety, the DNA was found to adsorb in a single kineticstep involving simultaneous adsorption, desorption and diffusion.100 Petrovykh and colleaguesmonitored the formation of DNA SAMs through FTIR and XPS, observing the initial adsorp-tion via the nitrogenous bases of the DNA onto bare gold, followed by chemisorption via thethiol. A log time dependence on the increasing coverage was observed and was rationalizedto be due to rate-limiting reorganization of the DNA SAM.96 Wackerbath and colleagues sawsimilar observations and observed reversible domain formation using in-situ STM measure-ments.101 AFM measurements performed by Erts and colleagues found that with increasingDNA coverage, the adsorbed DNA becomes more upright.98 Similar to alkanethiol SAM for-mation, self-assembly of SAMs containing longer DNA are slower due to the retardation of the“standing-up” of the DNA. This is a result of subsequent adsorption of other DNA molecules be-coming blocked by more nitrogenous bases non-specifically adsorbing to the gold surface.102Alternative methods of forming DNA SAMs have been proposed, such as with thiol-exchangeof a MCH SAM with thiol-modified DNA99,103 and a one-step codeposition, where the alka-nethiol passivator and thiol-modified DNA simultaneously adsorb onto bare gold from solu-tion.104–106 DNA SAMs made with the thiol-exchange method were characterized throughfluorescence microscopy103 and AFM99 and were found to have less DNA aggregates andnon-specific adsorption. Despite the existence of these alternative methods to make DNASAMs, the majority of research groups manufacture DNA biosensors using the conventionalapproach where the gold surface is immersed in a DNA solution followed by back filling theSAM with mercaptohexanol. As will be seen later, the process to make DNA SAMs plays arole in the resulting sensor performance.3.1.3 Applications of DNA SAMs as DNA BiosensorsThe most common application of DNA SAMs are nucleic acid sensors also known as DNAbiosensors. These DNA biosensors are one of many types of devices designed to detect andmeasure relevant biomolecules as a result of a disease or a medical condition. If the con-centration of these biomolecules deviate from normal values, then the presence of a diseaseor medical condition is confirmed. Since these biomolecules typically exist at low concentra-483.1. DNA Self-Assembled Monolayerstions, biosensors must have high sensitivity and low detection limits. Current biomoleculardetection strategies rely on sophisticated equipment specific to clinical laboratories. This isin contrast to the miniaturized and portable nature of DNA biosensors, which would provideeasily accessible diagnostics and therapeutics to the public.2 Multiplexed detection, sequenc-ing genes and measuring protein-DNA interactions can be done in parallel using nucleic acidmicroarrays. These microarrays are composed of micrometer sized islands of DNA adsorbedonto solid surfaces, each containing a different sequence, which enables detection of multi-ple biomolecules at the same time.107 In addition, the use of these nucleic acid arrays hasprovided a better alternative to slow Sanger sequencing.10,108Initial prototypes of DNA biosensors detected nucleic acids in a sample through hybridis-ation of a target analyte sequence to a complement probe sequence. Analytical signals weregenerated during this hybridisation event indicating the target sequence presence. Generally,these DNA biosensors consist of labeled probe sequence, which is more economical than la-beling the target sequence. Nucleic acid hybridisation exhibits differences at when occurringat the surface instead of in solution,109 which requires that additional parameters in the DNASAM must be controlled. Optimizing the DNA coverage is required in order to allow amplespacing at the surface for the probe DNA to hybridise.A review of select DNA biosensors manufactured through alkanethiol-modified DNA SAMson Au will be provided. Not every DNA biosensor will be detailed given the vast amount ofDNA biosensor research in literature.2 The DNA biosensors discussed will primarily focus onoptical and electrochemical sensors involving the probe sequence tethered to the surface.Fluorescence-basedDetection of nucleic acids in solution has involved the use of molecular beacons. A molec-ular beacon consists of probe DNA modified with a fluorophore on one end and a quencheron the other end. With no target present, the probe DNA is designed to be self complemen-tary and therefore forms a hairpin structure. The fluorophore is close to the quencher anddoes not fluoresce. In the presence of the target sequence, the probe sequence complementswith the target sequence, opening the hairpin structure and distancing the quencher from the493.1. DNA Self-Assembled MonolayersFigure 3.3: Schematic of a molecular beacon used to detect the presence of a target DNAsequence in solution. The fluorophore is quenched in the unbound state of the probe hairpinDNA. Upon hybridisation of the target DNA with the probe DNA, the fluorophore is separatedfrom the quencher and fluoresces.fluorophore which then is able to fluoresce (Figure 3.3).110Du and colleagues translated the molecular beacon to the surface where the hairpin probeDNA was immobilized onto a Au surface sputtered on quartz. Without target DNA present, thefluorophore was quenched by the gold surface. Upon hybridisation with the target sequence,the fluorescence signal increased as the fluorophore was no longer quenched by the sur-face (Figure 3.4a).111 The authors manufactured their sensor using a co-deposition approachwhere the DNA SAM was formed by exposing the Au surface to both DNA and mercapto-propanol for 2 hours. It was reported that the presence of non-specifically adsorbed DNA ledto significant background fluorescence. Surfaces were incubated for a short time to form lowcoverage DNA SAMs, which allowed sufficient spacing between the tethered DNA for hybridi-sation.15Rant and colleagues developed another fluorescence-based detection strategy for hybridi-sation involved an applied potential to actuate the probe DNA. With metal-mediated quenching,the fluctuating fluorescence intensity due to a modulated applied potential was used to monitorthe movement of the DNA.116 Hybridisation of the probe DNA with the target DNA resulted indsDNA. With the increased rigidity of dsDNA compared to ssDNA, an increased fluorescencesignal was observed (Figure 3.4b). This has been furthered for sensing proteins using ap-tamer DNA SAMs.117 The function of this sensor depended on the tethered probe DNA beingmobile when actuated by the potential. To achieve this, the DNA coverage for these sensorswere controlled through partial reductive desorption of the DNA SAMs which also eliminated503.1. DNA Self-Assembled MonolayersFigure 3.4: Example DNA biosensors approaches which use (a) fluorescence labelling withthe molecular beacon design, (b) potential-induced actuation with measurements, (c) redoxactive Ru(NH3)6Cl3, (d) Ferri/ferrocyanaide, (e) redox active intercalators, (f)(g)(h) covalentredox active labels. The molecules in these figures are not drawn to scale. Schematics weredrawn based on descriptions from various works.85,111–116513.1. DNA Self-Assembled Monolayersnon-specifically adsorbed DNA.118It is clear that these fluorescence-based biosensors were affected by non-specific adsorp-tion and required DNA SAMs with an optimal DNA coverage. Without reducing non-specificallyadsorbed DNA and optimizing the DNA coverage, alternative hybridisation detection strategieswere simultaneously being explored in order minimize the dependence on these factors.Electrochemical Labelling with DNA Groovebinders and Anionic RepulsionElectrochemical detection in DNA biosensors has been considered more favourable com-pared to optical detection due to the simplicity of electrochemical instrumentation over opticalinstrumentation and higher specificity of surface events; fluorescence from solution contributesto background signals whereas electrochemical signals only originate from events at the sur-face.119Redox active cations in solution can label the DNA through electrostatic interactionswith the negative charge on the DNA phosphate backbone. Steel and colleagues proposed amethod of electrochemically quantifying the coverage of DNA using Ru(NH3)3+6 (RuHex),120a redox active cation capable of binding to DNA grooves.121,122 The 1:3 stoichiometric ratioof Ru(NH3)3+6 bound to DNA bases meant that the RuHex at the surface was correlated withthe amount of DNA adsorbed. RuHex redox provided a method of detecting hybridisation, asthe increase in DNA surface concentration upon target and probe hybridsation resulted in anincrease in RuHex at the surface (Figure 3.4c).115,120 It should be noted that the binding ofRuHex to DNA decreases in the presence of sodium ions. Furthermore, the electron trans-fer rate is affected by the molecular orientation of the DNA and becomes slower with higherDNA SAM coverages.123 This method of detection, while simple and independent of the DNAsequence, is not as sensitive as other detection methods due to a background signal existingwhen probe sequence is adsorbed onto the surface without the target.First studied by Turcu and colleagues, electrochemical detection of DNA was proposedtaking advantage of the repulsion between the negatively charged DNA with an anionic re-dox couple: Fe(CN) 3 –6 / Fe(CN)4 –6 . When the target DNA hybridises to the adsorbed probeDNA, the increased negative charge at the surface prevents ferri/ferrocyanide migration to thesurface (Figure 3.4d). Changes to the kinetics of ferri/ferrocyanide redox were observed withEIS measurements.85,124,125 This approach, while instrumentally simple, was found to be influ-523.1. DNA Self-Assembled Monolayersenced by the orientation of the adsorbed DNA and signals varying based on characteristics ofthe SAM. Changes to the EIS signal were not only affected by the amount of DNA at the sur-face but also by the number of pinhole defects in the SAM. Higher concentrations and longerincubation times of the alkanethiol passivator (which is typically MCH), needed to remove thenon-specifically adsorbed DNA, resulted in decreased sensitivity to hybridisation.13Recently, Lam and colleagues combined the use of RuHex and ferri/ferrocyanide, providinga highly sensitive way of detecting target DNA. With ferri/ferrocyanide in solution enhancedthe current by reoxidizing the RuHex which was being electrochemically reduced during themeasurement resulting in a higher signal to noise ratio.126High background signals are unavoidable when using RuHex and ferri/ferrocyanide whichare both indiscriminately sensitive to the probe DNA and target DNA. The lack of selectivelabeling also means signal changes can result from the presence of non-specifically adsorbedDNA. Electrochemical detection where a signal is generated only when dsDNA is formed ispreferred, motivating research of redox-active labels which can distinguish dsDNA from ss-DNA.Electrochemical Labeling with DNA IntercalatorsRedox-active compounds that intercalate specifically with dsDNA have been used to detecthybridisation. Compounds, such as 2,6-anthraquinonedisulfonic acid, Ir(bpy)(phen)(phi)3+ anddaunomycin to name a few, were demonstrated by multiple groups to preferably bind to dsDNAover ssDNA.10 When sensing target DNA, an electrochemical signal is generated only whenthe target sequence had hybridized to the probe sequence (Figure 3.4e). The use of interca-lators for specific detection of the target sequence and differentiation from targets containingsingle basepair mismatches was explored by Kelley, Boon and colleagues.127–129 For thesesensors, DNA SAMs with high coverage were preferred to promote only specific hybridsationof the probe and target sequence. Hybridisation between mismatched strands was reduced inhigh coverage DNA SAMs. While manufacturing the DNA SAMs for these sensors, no alka-nethiol passivator was mentioned. As it is typically used to remove non-specifically adsorbedDNA,89,91 non-specific adsorption was likely not considered and therefore not addressed.However, later studies did find that intercalators could also bind to single stranded DNA,533.1. DNA Self-Assembled Monolayersreducing some advantages of intercalators over detection with RuHex and ferri/ferrocyanide.In addition, certain intercalators were found to bind to specific sequences, making it difficult tofabricate sensors containing a different sequence.The sequence the intercalators recognizedwere short, meaning sensors where the probe and target sequences were longer exhibitedlower signal to noise ratios.10 Similar to detection using ferri/ferrocyanide, detection with inter-calators depended on characteristics of the DNA SAM. Binding of the intercalator with dsDNAdepended on the orientation of the DNA in the SAMs and the extent of hybridisation betweenthe probe and target DNA. The signal was also seen to be affected by the sequence dependentcharge transfer properties of DNA.10,109,130 These factors would significantly affect the perfor-mance of a commercialized nucleic acid sensor if signals were inconsistent across differentDNA SAMs. An electrochemical labelling method independent of the characteristics of theDNA SAM but still specific to the presence of target DNA was found using covalently attachedelectrochemical labels.Electrochemical Labeling with Covalent LinkersTo avoid complications from using groove-binders or intercalators, DNA sensors containingredox-active labels covalently attached to the probe were developed. Labels, such as osmiumtetroxoide, ferrocene, anthraquinone,and methylene blue had higher stability compared to in-tercalators. Using covalent labels also allowed for consistent control of label location on theprobe DNA.10,112 Methylene blue has been primarily used as a covalent label, exhibiting higherstability compared to ferrocene in blood matrices.131Sensors using these covalent labels, named E-DNA sensors, have been favourable overintercalators and groovebinders as no added reagents are needed when sensing.112,119 Thesensor constructed imitated the optical sensor used by Du and colleagues.111 Signals weregenerated when the redox-active label attached to the distal end of the adsorbed probe DNAwas able to access the electrode surface while the probe DNA was in a hairpin configuration.Hybridisation with the target strand would open the hairpin and reduce the accessibility ofthe redox-active label to the surface and reduce the electrochemical signal. With this mech-anism, the probe DNA must undergo a conformation change in the presence of the target543.1. DNA Self-Assembled Monolayersstrand; a disadvantage that did not apply to DNA sensors using redox active intercalators orgroove-binders. Furthermore, Valée–Bélisle and colleagues noted that careful choice of theprobe DNA sequence is needed in order to prevent the hairpin probe opening when no targetstrand is present (which would lead to high background signals) without causing the hairpinstate to be more energetically favourable than the dsDNA state (which would prevent targetbinding).132The first of these E-DNA sensors was proposed by Fan and colleagues. Here,ferrocene was covalently attached to the end of the probe DNA as depicted in Figure 3.4f.112This sensor, coined the “signal-off” sensor, was improved upon with the proposed “signal-on”sensor. In this approach, the probe DNA was initially hybridized to a complement strand con-taining the covalently linked redox label. With the rigidity of dsDNA, the redox-active label washindered from accessing the surface and no electrochemical signal was generated. In thepresence of target DNA, the probe DNA hybridized with the target DNA and in doing so, theredox labeled complement partially dehybridises from the probe DNA allowing the redox labelto access the surface and generating a higher electrochemical signal (Figure 3.4g).114Biomolecular Detection with Aptamer SensorsUsing aptamers as the probe DNA in E-DNA sensors have been used to detect otherbiomolecules besides nucleic acids. Aptamers are DNA with a carefully chosen sequencewhich specifically bind to a particular biomolecule.6 Aptamers that undergo a change in con-formation upon binding to a biomolecule are preferred in the E-DNA sensors design.9,133 Forcontinuous monitoring of these biomolecules in a flow-through apparatus, the binding betweenthe biomolecule and the aptamer needs to be reversible.7Thrombin, a protein that causes blood clotting, has been demonstrated to specifically bindwith an aptamer that forms a secondary structure (a G-Quadruplex) upon binding. In usingthis aptamer for a E-DNA sensor, Xiao and colleagues took advantage of the the secondarystructure forming. When the probe DNA formed the G-Quadruplex, the redox active marker atthe distal end of the probe DNA was restricted from accessing the surface, reducing electrontransfer of the methylene blue to the surface (Figure 3.4h).113 Generally, secondary structuresoccupy a larger area on the surface in comparison to ssDNA. Thus high DNA coverages willrestrict the conformational change, therefore control of the local DNA environment is required553.1. DNA Self-Assembled Monolayersto optimize the sensitivity of aptamer E-DNA sensors.1343.1.4 Current Challenges with DNA SAM ApplicationThe mentioned DNA sensors are only a few of the ongoing efforts to develop a DNA biosen-sor, but show clear examples of the research progress. Despite substantial research progress,commercialization of DNA biosensors as a point-of-care diagnostic device has not been re-alized.2 Furthermore, only four FDA-approved microarray based diagnostics have been mar-keted for use by the health care professionals.135 This section will detail the challenges andlimitations that have slowed the application of DNA SAMs as functional biosensors for use indiagnostics.When detecting nucleic acids in a real sample, complications arise during the detectionof a target analyte in a complex matrix. Most initial tests for DNA sensors have been donewith samples that only contain the target sequence. A real sample would consist of a matrixcontaining non-complementary DNA, which would compete for binding with the probe DNAat the surface and reduce the sensitivity of the DNA sensor. Amplification of the short targetsequences can be done but adds complications to instrumentation and adds to sample prepa-ration time.10 It becomes increasingly difficult to detect target molecules in biological matriceswhen the DNA SAM becomes contaminated through biofouling. Though oligo(ethylene glycol)terminated thiols prevent contamination of the surface from proteins, the electrode becomeselectrically insulating, which can affect electrochemical sensing.136 Few studies have beendone detecting biomolecules in real biological samples, along with a comparison to currentclinical methods of detection. Thus, DNA biosensors are unable to compete with and replacecurrent biomarker detection systems used in clinical diagnostics.137As mentioned in the previous section, inconsistencies between replicate DNA SAMs pre-pared in the same manner have been cited to result in DNA sensors with irreproducible sig-nals.10,13,109,130 The sources of these inconsistencies have not been fully identified though itis speculated to be a result of the non-homogeneous nature of surfaces compared to solu-tion.12,138 Sophisticated data processing methods have been proposed in order to extract in-formation from the inconsistent data.138 Alternatively, identifying and addressing the cause ofthese irreproducibilities have been done in recent years. Contaminants have been shown to in-563.1. DNA Self-Assembled Monolayerstroduce inconsistencies by interfering with the alkanethiol-modified DNA adsorption to the sur-face.139 Inconsistencies are also introduced by the uncontrollable presence of non-specificallyadsorbed DNA.140 Non-specifically adsorbed DNA can be a source of false positive signalsdepending on the labeling method. Meanwhile, slow removal of non-specifically adsorbedDNA over time or with rinsing leads to decreasing signal with subsequent uses of the DNAsensor.104,141 Non-specifically adsorbed DNA was not detected during initial characterizationof DNA SAMs made with the manufacturing procedure proposed by Herne and Tarlov.15,89,100However, attempts to use DNA SAMs for nucleic acid sensors revealed the lingering presenceof non-specifically adsorbed DNA which reduced sensor performance.104,141,142Varying the DNA coverage and therefore the DNA local environment can affect the mech-anism of probe-target hybridisation in the DNA sensor. High coverage DNA SAMs stericallyhinder target and probe interaction at the surface. This is observed in Figure 3.5 where thehybridisation efficiency, which is the percentage of probe DNA hybridised with target DNA,decreases with increasing DNA coverage. Lowering the DNA coverage would in turn lead toa decrease of probe DNA at the surface and therefore a reduced signal.15,143 Depending onthe design of the sensor, a high coverage can actually be beneficial to the sensor sensitivity.For example, for the e-DNA sensor proposed by Fan and colleagues,112 a high coverage DNASAM prevents the redox-active label attached in the dsDNA (containing both the probe andtarget DNA ) from accessing the surface, reducing the chance of a false negative signal.134On the other hand, in sensors where the DNA undergoes a conformational change upon bind-ing to the target, the local environment or spacing between the probe DNA would need to besufficient for this to occur. Therefore, a lower but optimal coverage is required for the largestsignal change when detecting the target sequence or biomolecule.144Inconsistent DNA coverage over time, resulting in a loss of signal from the DNA sensor,can be due to desorption of probe DNA, originating from a defect in the DNA SAM.145 Increas-ing the stability of DNA SAMs and minimizing the creation of such defects in the DNA SAMis of utmost importance in order to reduce this loss. While DNA SAMs made can exhibit awide variety of non-ideal behaviour which lead to irreproducible signals, the work in this the-sis specifically addresses inconsistent DNA SAM formation as a result of a heterogeneousdistribution of DNA coverages across the surface.573.1. DNA Self-Assembled MonolayersFigure 3.5: Hybridisation efficiencies measured using surface plasmon resonance for DNASAMs with varying coverages. Increasing DNA coverage results in lower hybridisation ef-ficiency. Adapted from Peterson and colleagues.15 Copyright (2001) with permission fromOxford University Press.Heterogeneous distribution of the adsorbed DNA, as a result of DNA aggregation or sub-strate effects, introduce further inconsistencies in the formation of DNA SAMs. For example,a heterogeneous DNA SAM could contain domains each with a different surface coverage. Ahighly heterogeneous SAM would contain a wide variety of DNA coverages whereas a homo-geneously distributed DNA SAM ideally consist of DNA with identical local environments or aconsistent DNA coverage over the surface. As mentioned previously, an optimal coverage isrequired for a DNA biosensor to have high hybridisation efficiency without compromising withsensitivity. For a DNA SAM which the DNA is homogeneously distributed, the overall DNAcoverage would be a good indication of the hybridisation efficiency. In the extreme case of aheterogeneous DNA SAM where there are only regions of extremely high surface coverage(where hybridisation efficiency is low) and regions of extremely low surface coverage (wheresensitivity is low), a DNA sensor made with this SAM would not be effective. Furthermore,when comparing two DNA SAMs with the same overall coverage but with different degrees ofheterogeneity, the hybridisation efficiencies on these two sensors will be different, leading toDNA biosensors yielding inconsistent results. By reducing DNA heterogeneity, optimizing thehybridisation efficiency can be accomplished by controlling the DNA coverage only. However,with DNA SAMs typically prepared on polycrystalline gold substrates,146 heterogeneities in theSAM due to the influence of surface crystallographies are unavoidable.12 Not only does the583.1. DNA Self-Assembled MonolayersDNA coverage have to be controlled but also the number of defects and the degree of het-erogeneous DNA coverage need to be reduced to increase the SAM stability. One proposedmethod of achieving this is by assembling DNA SAMs using potential-assisted deposition, alsoknown as electrodeposition.3.1.5 Potential-Assisted Formation of DNA SAMsControlling the DNA coverage is crucial for optimizing DNA biosensor performance. Vari-ables affecting DNA coverage in SAMs include changing incubation time, concentration ofDNA in solution or ionic strength during DNA deposition.15,89,91 While manipulating one ofthese variables effectively changes the DNA coverage, there have been recent endeavors inusing an applied potential to the electrode to control the coverage.15,147 It has been reportedthat using an applied potential enables consistent formation of alkanethiol SAMs, which hadbeen lacking when forming alkanethiol SAMs with conventional methods.14 An added advan-tage to using potential control during self-assembly is the ability to tailor the coverage on anarray of electrodes while using the same deposition solution.148,149To understand the role of potential during the self-assembly process, the open circuit po-tential (OCP) was measured during the adsorption of alkanethiols on gold. Upon introductionof the alkanethiol in solution, the OCP became more negative due to donation of one electronper alkanethiol to the gold surface followed by a slow positive discharge.150 From these obser-vations, it was rationalized that applying a positive potential would enhance the adsorption ofalkanethiols by facilitating a faster discharge.14,150Ron and Rubinstein along with Paik and colleagues proposed three explanations for theenhanced adsorption of alkanethiols at the positive potential : (1) electrochemical oxidation ofthe gold surface, (2) increasing surface sites available by inducing standing up behaviour ofthe alkanethiols, or (3) removal of a less ordered portions of the monolayer therefore allowingonly a stable SAM.151,152Ma and Lennox deposited alkanethiol SAMs at different deposition potentials from an alka-nethiol solution in LiClO4/ethanol for 10 minutes with the resulting SAMs characterized withEIS measurements. Alkanethiol SAMs immobilized while applying a potential more positivethan the OCP adsorbed at a faster rate and contained less defects compared to SAMs made593.1. DNA Self-Assembled Monolayersat negative potentials or at OCP.14 Similar contact angles were observed, however, betweenall the layers after 10 minutes, suggesting the properties of the surface and the coverage to besimilar with or without an applied deposition potential. At the positive potentials applied duringthese alkanethiol deposition, it was also not possible for the alkanethiol (RSH) to be oxidized orfor Au oxide formation to occur at the deposition potentials. It was concluded that the potentialcauses rearrangement and reorganization of the forming SAM which appeared as a SAM withdefects.14Similar results were reported by Peterson and colleagues when depositing thiol-modifiedDNA onto gold while applying 0.3V vs. Ag/AgCl in 1 M NaCl. The thickness of the DNASAM, measured using surface plasmon resonance, grew significantly faster during depositioncompared to the DNA SAM deposited without any applied potential.15 Recently, Kroener andcolleagues have demonstrated that applying potentials during deposition enables preferentialadsorption of large 100 basepair alkanethiol-modified DNA origami structures over 40 basepairnon-alkanethiol modified DNA.153 While this work showed that potential-assisted DNA deposi-tion methods enhanced DNA adsorption, the effect of potential on heterogeneous features inthe DNA SAMs had not been addressed.In DNA SAMs formed without potential-assisted deposition, DNA aggregation has been ob-served through imaging.103 If these heterogeneous features are pertinent to DNA SAMs madeusing potential-assisted deposition, they will have to be examined through a similar imagingtechnique. Future DNA biosensors will be prepared using sputtered gold films,146 made upof polycrystalline surfaces containing many small grains of various orientations.73 DNA SAMsadsorbed on top of these will be influenced by the various surface crystallographies, which isknown to affect the structure and adsorption behaviour of alkanethiol SAMs.11,19 Studies ofthe potential-assisted deposition or modification of a DNA SAM have not been studied acrossdifferent surface crystallographies. With the knowledge of different pzc’s for the different sur-face crystallographies,20,21,25,71 investigating the potential-assisted formation of DNA SAMs for603.2. in-situ Electrochemical Fluorescence Microscopy to investigate DNA SAMsdifferent surface crystallographies with a systematic surface analytical technique is important.3.2 in-situ Electrochemical Fluorescence Microscopy toinvestigate DNA SAMsA method used to investigate heterogeneities in DNA SAMs is in-situ electrochemical flu-orescence microscopy imaging (iSEFMI). This technique combines electrochemical measure-ments on metal electrodes with simultaneous fluorescence microscopy imaging and has beenused to study adsorbates at metal surfaces. A description of iSEFMI will be provided alongwith a brief description of studies involving characterization of self assembled monolayers,including DNA SAMs.3.2.1 Combining Electrochemistry with Fluorescence MicroscopyiSEFMI measurements originate from spectroelectrochemical setups where the fluores-cence intensity of fluorophores immobilized at or near metal surfaces was monitored withchanging the applied potential. Lateral-resolved spectroelectrochemical microscopy was firstdeveloped by Bizzotto and Pettinger in order to study the properties of a octadecanol layer con-taining fluorophore labeled molecular probes on a gold electrode.154 The most recent iterationof iSEFMI uses an inverted epi-fluorescence setup where the fluorescent sample is excitedby light through the same objective which the resulting fluorescence is collected. Filtering theexcitation and emission light is done through a filter cube containing a dichroic mirror and exci-tation and emission filters chosen for detecting the fluorescence of a particular fluorophore.16The fluorescence intensity correlates with the concentration of fluorophore at the surfaceand decreases with metal-induced fluorescence quenching. As discussed in section 2.3.2,the distance the fluorophore is from the surface changes the fluorescence intensity, whichcan occur as a result of an applied potential to the metal surface. Information on the localenvironment in which the fluorophore labeled molecules reside can be obtained from this in-formation. Simultaneous electrochemical measurements (e.g. capacitance, CV, CC) can bedone to provide additional characterization of the metal-solution interface and an averagedmeasurement of the entire electrode surface. This enables correlation of the fluorescencemeasurements, which represent the movement of fluorescently labeled molecules at the sur-613.2. in-situ Electrochemical Fluorescence Microscopy to investigate DNA SAMsface, with the electrochemical response. Overall, the larger amount of data generated whencombining electrochemical measurements with fluorescence measurements can be beneficialfor studying fluorescently-labeled alkanethiol SAMs on gold electrodes.163.2.2 iSEFMI to Study Alkanethiol Self-Assembled Monolayers on MetalElectrodesImaging alkanethiol SAMs adsorbed onto metal surfaces provides a way to examine theirheterogeneity. This has been shown previously in scanning probe microscopy studies us-ing AFM and STM to evaluate alkanethiol SAM formation.22,66–68 iSEFMI is another imagingmethod which has been used to evaluate the conformation, organization and electrochemicalresponse of alkanethiol SAMs.72,73,155Shepherd and colleagues imaged fluorescently labeled alkanethiol SAMs on polycrystallinegold electrodes while applying negative potentials. When reductive desorption of the SAM oc-curred, an increase in fluorescence intensity was observed, due the metal-induced quenchingdiminishing as the fluorophore moved away from the surface. The fluorescence intensity sub-sequently decreased as the fluorophore diffused into the electrolyte. Since imaging allows si-multaneous characterization on different regions of the electrode, the applied potentials whichreductive desorption on certain regions of the SAM occurred, could be recorded.73 A corre-lation between the reductive desorption potential and the surface morphology along with theunderlying gold surface crystallography was observed, agreeing with previous research on theinfluence of surface crystallography on the pzc and the reductive desorption potential.20,21,25,71Musgrove and colleagues examined oxidative desorption of alkanethiol SAMs on polycrys-talline Au electrodes using iSEFMI. An increase in fluorescence was observed while applyingpositive potentials as the adsorbates desorbed from the surface then diffused away.72 In com-parison to reductive desorption, oxidatively desorbed molecules did not immediately leave thesurface. It was unclear if the oxidative desorption potential was dependent on the surfacecrystallography. However, the authors observed the same oxidative desorption potentials onsurfaces with similar morphology.Casanova-Moreno and Bizzotto examined the reductive desorption of fluorescently labelledalkanethiol SAMs immobilized onto Au microelectrodes tilted up to 9° with respect to the hori-623.2. in-situ Electrochemical Fluorescence Microscopy to investigate DNA SAMszontal plane. The movement of the reductively desorbed alkanethiols was found to be relatedbe affected by the microelectrode (the working electrode) with respect to the counter electrodeand buoyant forces from hydrogen generated from the simultaneous water-splitting process.155Thiol SAMs containing charged molecules proved to be more sensitive to applied poten-tials. Alkanethiol DNA SAMs, with its negatively charged phosphate backbone, has beenshown by Kaiser and Rant to reorient due to the potential at the surface, provided that the ionicstrength in solution was low enough.88 The orientation of fluorescently labeled DNA could beevaluated based on the distance dependent metal-quenching of fluorophore. Positive poten-tials attract the DNA to the metal surface, and the fluorescence intensity decreases. Negativepotentials repels the DNA, which “stands-up” and the fluorophore is further from the surface in-creasing the fluorescence intensity.156 The changes in the fluorescence intensity with positiveand negative potentials were found to correlate with the tethered DNA spacing at the surfacewhich depended on the DNA coverage.157Fluorescence images of DNA SAMs were obtained by Murphy and colleagues using iSEFMIwhere aggregates of DNA were revealed.103 Instabilities in the DNA SAM, observed as thedesorption of DNA occurring at potentials less negative than reductive desorption, revealedthe presence of non-specifically adsorbed DNA.103,140 With iSEFMI, the factors which affectformation of DNA aggregates and non-specifically adsorbed DNA could be identified. More-over, it was found that DNA SAMs made via the thiol-exchange of DNA with a mercaptohexanolSAM resulted in less DNA aggregation and had higher stability compared to DNA SAMs madeusing the conventional DNA immobilization with mercaptohexanol back-filling method or withthe co-deposition method.103,158Heterogeneities in DNA SAMs have been investigated through scanning probe microscopytechniques. However, this was done with some difficulties due to the size of the DNA as wellas the retained water and salt in the SAM. Erts and colleagues initially investigated diluteDNA monolayers using AFM,98 heterogeneities of DNA SAMs at low coverages have beeninvestigated using AFM by the Ye group99,159 and STM was used to characterize DNA SAMslabelled with RuHex by Grubb and colleagues.160,161 The advantage of using iSEFMI to inves-tigate DNA SAMs include the ability to investigate the DNA SAM in an aqueous environmentas well as the ability to investigate DNA SAMs with different coverages. It is clear that using633.3. Contributions of the Research Presented in this ThesisiSEFMI is beneficial for detection and remediation of non-idealities in DNA SAMs.123.2.3 Observed Surface Crystallography Dependence for Thiol SAMsElectrochemical studies and initial iSEFMI studies have indicated that reductive desorptionof SAMs depends on surface crystallography.71,73 However, Yu and colleagues developed amore systematic method of evaluating the influence of surface crystallography on the reduc-tive desorption potential by examining fluorescently labeled alkanethiol SAMs immobilized onspherical single crystal bead electrodes. These electrodes, which were made by melting andslowly cooling a gold wire, enabled a way to optically image all surface crystallographic ori-entations simultaneously.19 With the predetermined spatial arrangement of each surface crys-tallography previously mapped on the single crystal bead electrodes, properties of alkanethiolSAMs, thiol-modified peptide SAMs and alkanethiol DNA SAMs on the single crystal beadelectrodes were correlated with the underlying surface crystallography. Additionally a self-consistent comparison of SAM formation across different surface crystallographies could bedone, where all the conditions during involved in SAM formation (e.g. adsorbate concentration,electrolyte composition, incubation time) would be the same across all surface crystallogra-phies. In forming the DNA SAMs through the thiol-exchange of DNA with a mercaptohexanolSAM, a variation of DNA coverage was observed on different surface crystallographies.19,162Reductive desorption potentials were also correlated with the underlying surface crystallogra-phy establishing this relation on more surface crystallographic orientations other than Au(111),Au(100), Au(110) and Au(210).193.3 Contributions of the Research Presented in this ThesisWithin the literature, it is evident that there are challenges in creating DNA SAMs for the pur-pose of DNA biosensors. Multiple factors (e.g. concentration, deposition time, ionic strength)that are involved in the manufacturing process need to be manipulated in order to make aDNA sensor with optimal sensitivity and stability. To accommodate the wide range of DNAbiosensor architectures that exist, a facile method of manipulating the DNA coverage as wellas the local environment is sought after. Additionally, previous studies have shown that the ir-reproducibility of these DNA SAMs affects their reliability. With iSEFMI analysis of DNA SAMs,643.3. Contributions of the Research Presented in this Thesisthe heterogeneity of DNA SAMs had previously been revealed to be affected by the substrate,which is in turn affected by the surface crystallography. Previous studies have shown thatpotential-assisted deposition of DNA can create DNA SAMs in less time compared to the con-ventional manufacturing process with more control and fewer defects. However, no informationhas been attained regarding potential-assisted deposition of DNA SAMs on different surfacecrystallographic features.The thesis will detail potential-assisted DNA SAMs on spherical single crystal Au bead elec-trodes analyzed using iSEFMI to reveal the influence of surface crystallography, the concen-tration of thiol-modified DNA in the deposition buffer, and deposition potentials on the resultingDNA SAM. With the ability to observe the influence of surface crystallography with iSEFMI,controlling the DNA coverage on certain surface crystallographic regions is possible. Theseresults will shed light on previous potential-assisted deposition studies and on the mechanismby which potential influences DNA SAM formation for a range of surface crystallographies.Based on the work in this thesis, an optimized procedure (with a given deposition buffer,applied potential, and deposition time) can be developed in order to form DNA SAMs con-sistently where the local environment around each DNA is the same throughout the SAM. Inthe following chapters, studies will be detailed demonstrating how exactly the applied poten-tial can influence both heterogeneity and DNA coverage in the DNA SAM as well as the localenvironment for each adsorbed DNA. With applied potential being able to control the DNAcoverage, optimization and tailoring of DNA SAMs to the needs of any biosensor motif willbe possible. Additionally, DNA SAMs can be made in a shorter time span (< 24 hours) withpotential-assisted deposition, shortening the manufacturing time and improving mass produc-tion of DNA sensors. This will lead to DNA sensors with higher reproducibility paving the wayfor reliable DNA sensor to become commercialized.65Chapter 4ExperimentalA summary of the common experimental approaches, materials and instrumentation used inChapters 5,6 and 7 are described here. Specific experimental procedures and measurementparameters used for a particular project will be detailed in individual chapters.4.1 Materials4.1.1 Gold Substrates and CleaningAll experiments were performed on gold spherical single crystal bead electrodes enabling aself-consistent comparison of DNA SAMs on different surface crystallographies. These elec-trodes were made from a gold wire (1mm diameter, 99.999%, Alfa Aesar) that was repeatedlymelted with a butane torch and cooled until a single crystal was obtained. Single crystal for-mation is confirmed by the presence of four relatively large Au(111) facets, or flat regions,arranged in a square surrounding a smaller Au(100) facet on the surface (Figure 4.1b). Theatomic model of a single crystal bead electrode shown previously in Figure 2.3 and Figure 4.1ashows this orientation of surface crystallographies that is inherent to gold single crystal beadelectrodes. The spherical single crystal bead electrodes used typically have a diameter of 1.5to 2 mm. Surface area measurements were done while applying -0.8V/SCE to the electrodein 0.1 M KClO4 and measuring the capacitance.163 The area is calculated based on the nor-malized capacitance of a gold multicrystalline surface(17.5 μF/cm2), which is the case in anelectrolyte not containing specifically adsorbing ions.The surfaces of the Au single crystal bead electrodes required cleaning prior to the deposi-tion of monolayers. Between each cleaning procedure, substrates were rinsed using Milliporewater (>18 MΩ resistivity), purified with a MilliQ® Integral 5 water purification system. Single664.1. MaterialsFigure 4.1: a) Projection of the surface of a single crystal bead with certain surface crystallo-graphic features labeled. b) Brightfield image of a single crystal bead electrode with the samesurface crystallographies. Only Au(111) and Au(100) are visible in brightfield mode based ontheir physical features.crystal Au substrates were initially exposed to aqua regia (HNO3 and HCl 1:3 volume ratio)for 15 minutes then remelted. Electrochemical cleaning of the electrode was done with cyclicvoltammetry in approximately 0.1 M H2SO4 (-0.4 V and 1.45 V at 50 mV/s) until the cyclicvoltammmogram did not change. Immediately prior to deposition of monolayers, the electrodewas flame annealed using a butane torch.4.1.2 Modifying Au substratesImmediately after cleaning, DNA SAMs were formed on the single crystal bead electrodes.This was done by immersion of the clean substrate into a series of solutions while applyingpotential during immersion in the DNA solution. The DNA deposition was done in a DNA so-lution in an eppendorf tube where the Au electrode was immersed next to a glass salt bridgeconnected to a reference electrode. Details involving this setup and the electrochemical pro-cedures which took place during the DNA deposition will be detailed in later chapters.DNA was purchased from Integrated DNA Technologies, Inc. with a HO(CH2)6SS(CH2)6modification on the 5’ end and a fluorophore AlexaFluor®488 on the 3’ end. The 30 basepair DNA sequence (5’CTG TAT TGA GTT GTA TCG TGT GGT GTA TTT 3) does not formsecondary structures as confirmed by oligocalc and mfold.164,165 DNA solutions were storedin a 10 mM Trisaminomethane (TRIS) pH 7.5 buffer (containing TRIS base and TRIS HCl,Bioperformance >99.0%, Sigma Aldrich) diluted with Millipore water and stored at -20 °C inan eppendorf tube. The HO(CH2)6SS(CH2)6-R moiety 20 μM DNA solution was reduced to674.1. Materials-50-40-30-20-1001020-0.5 0.0 0.5 1.0 1.5 I (µA)E (V/SCE)Figure 4.2: Cyclic voltammogram of a Au spherical electrode in approximately 0.1 M H2SO4cycled between -0.40 V and 1.45 V vs. SCE at 50 mV/s scan rateHS(CH2)6-R for 3 hours using 0.01 M tris(2-carboxyethyl)phosphine-hydrochloride or TCEP(>98% Sigma Aldrich) with 0.01 M KOH (99.99% semiconductor grade Sigma Aldrich). DNAwas then purified using a General Electric MicrospinTM G-50 column, pre-rinsed with pH 7.5TRIS buffer then stored at -20 °C prior to dilution in the immobilization buffer. The HS(CH2)6-Rmoiety remained unoxidized (to HO(CH2)6SS(CH2)6-R) for at least three weeks in the freezerwhich was confirmed through polyacrylamide gel-electrophoresis.Generally, DNA SAM formation in this work is done using two different methods: Thiol-exchange of an alkanethiol SAM with DNA and DNA adsorption on bare/clean Au.For thiol-exchange of an alkylthiol SAM with DNA, the cleaned electrode is immersed first in a solutioncontaining 1 mM mercaptohexanol (MCH ,99% Sigma Aldrich) for 30 minutes. This solutionwas diluted in HPLC grade methanol (MeOH, Fisher Scientific) from a 10 mM stock also dilutedin methanol. For long-term storage, MCH was diluted to 50 mM solution in methanol at -20°C for up to one month. The Au electrode was then rinsed with MeOH then Millipore waterthen immersed into a DNA solution diluted to 1 μM or less in a pH 7.5 immobilization buffer. Inlater experimental chapters, the duration of this DNA deposition step, the composition of theimmobilization buffer (IB) and the apparatus which allows potential to be applied during thisDNA deposition will be detailed. Following DNA deposition, the DNA SAM was rinsed with684.2. CharacterizationMillipore water and stored in IB overnight.For DNA adsorption on clean Au, the cleaned gold electrode was immersed in the a DNAsolution diluted to 1 μM or less in an aqueous pH 7.5 immobilization buffer. The electrode-position apparatus used was identical to the one used during thiol-exchange and DNA wasprepared and purified in the same manner as stated above. Again, the duration of this DNA de-position, the composition of the immobilization buffer and the electrodeposition apparatus willbe detailed. Following DNA deposition, the electrode was rinsed with Millipore water then im-mersed in a 1 mM mercaptohexanol (MCH) solution for 90 minutes. The MCH solution was di-luted in the aqueous immobilization buffer from a 10 mM MCH stock solution in methanol. AfterMCH modification the gold electrode was rinsed in Millipore water and stored in IB overnight.4.1.3 Electrolyte Solutions for Electrochemical MeasurementsIn the next section, the electrochemical methods used to characterize DNA SAMs will be de-tailed. Electrolyte solutions for these electrochemical measurements were always made withMillipore water. Ruthenium Hexaammine (RuHex) electrochemical measurements to charac-terize DNA coverage were performed in a 10 mM TRIS pH 7.5 buffer containing 2.5 μM, 5μM or 50μM of Ru(NH3)6Cl3. EIS measurements with ferri/ferrocyanide measurements weredone in 10 mM phosphate buffer pH 7.5 containing 20 mM Na2SO4, 5 mM K3Fe(CN)6 and 5mM K4Fe(CN)6. For electrochemical fluorescence microscopy characterization, the electrolyteused contained a 10 mM TRIS buffer pH 7.5 with 10 mM KNO3.4.2 Characterization4.2.1 Instrumentation and ApparatusElectrochemical cleaning, DNA electrodeposition and electrochemical measurements not in-volving fluorescence imaging were done with the use of a Autolab potentiostat (PGSTAT30)and the software interface NOVA 1.11. The following set-up is not used for the DNA depositionand only for electrochemical cleaning and other electrochemical measurements on the sin-gle crystal bead electrode. The Au electrode was connected as the working electrode with a694.2. CharacterizationFigure 4.3: Typical electrochemical apparatus used for substrate cleaning and electrochemicalcharacterization. A saturated calomel electrode resides in a separate container containingsaturated KCl, is connected via a salt bridge to the electrolyte solution where the platinumcounter electrode and the Au substrate working electrode are immersed.platinum wire as a counter electrode (CE). The reference electrode (RE), a saturated calomelelectrode (SCE), was connected via a glass salt bridge as seen in the apparatus shown inFigure 4.3.Electrochemical measurements coupled with fluorescence imaging were done using a FHIpotentiostat. An epi-fluorescence Olympus IX70 microscope similar to that described in Shep-herd and Bizzotto166 is used to image DNA SAMs on Au single crystal bead electrodes within aspectroelectrochemical cell containing electrolyte with a Pt wire counter electrode and a teflonstopcock salt bridge containing a saturated calomel electrode reference (Figure 4.4). The beadelectrode was mounted at a consistent distance (~ 0.5 cm)from the window at the bottom ofthe spectroelectrochemical cell. Samples were illuminated with a Hg Arc 200 W Lamp (Xcite®Exacte w/ EXFO closed loop feedback) and images were taken at 5x magnification (NA=0.13)using a Photometrics Evolve® 512 electron multiplying charge-coupled device digital camera(512x512 pixels).Fluorescence was measured using a filter cube which was composed of an excitation filter,a dichroic mirror and and an emission filter chosen for detection of AlexaFluor®488 (Figure4.5). Light from the lamp was filtered through the excitation filter (450-490nm), then reflectedoff a dichroic mirror towards the sample. Fluorescence emission from the sample passed704.2. CharacterizationFigure 4.4: For electrochemical fluorescence microscopy, the Au substrate in a spectroelectro-chemical cell with a Pt coil electrode and a teflon stopcock salt bridge containing a saturatedcalomel electrode. The cell is mounted above an epi-fluorescence microscope where fluores-cence images are taken.through the same filter cube transmitting through the dichroic mirror (transmission >495 nm)and through an emission filter (500-550nm) before data collection in the camera.4.2.2 in-situ Electrochemical Fluorescence MicroscopyAll DNA SAMs formed on Au single crystal bead electrodes were evaluated using in-situ elec-trochemical fluorescence microscopy imaging (iSEFMI). An alternating potential step profilewhere potentials between -0.4V/SCE and 0.35V/SCE (Figure 4.6) were applied to the DNASAM. A fluorescence image was taken at each step with 3 second exposure time and EM gainof 50. The potential step profile was applied three times to verify that the measurements madewere reproducible and to confirm the lack of photo-bleaching and hysteresis. Photo-bleachingis minimized due to the low quantum yield of the fluorophore as a result of quenching by themetal surface. Background images were obtained at the same exposure time and EM gain af-ter applying -1.4V/SCE to reductively desorb the DNA SAM until no fluorescently labelled DNAis present on the electrode surface. The fluorescently labelled DNA once desorbed dispersesinto the electrolyte and assumed to no longer contribute to any measured fluorescence.714.2. Characterization(a)0255075100Rel. Int. (%)Exc. and Em. Spectra0255075100Rel. Int. (%)Excitation Filter0255075100Rel. Int. (%)Excitation Filter0255075100 300  400  500  600  700Rel. Int. (%)λ (nm)Dichroic Mirror(b)Figure 4.5: (a) Schematic of the filter cube that separates excitation and emission in the in-verted epi-fluorescence set up. (b) The excitation (blue) and emission (green) spectra for thefluorophore AlexaFluor®488 is shown as well as the transmission spectra for the excitationfilter, the emission filter and dichroic mirror used in the fluorescence microscopy configuration.Figure 4.6: The alternating potential profile applied during iSEFMI. At each potential step afluorescence image was taken.724.2. CharacterizationFluorescence images were analysed using ImageJ 1.48V software.167 A brightfield imageis initially taken with no additional light source so as to locate the Au (111) and Au (100)facets, which appear as flat surfaces with ridged edges (Figure 4.7). The large area of thesesurface crystallography is due to their low surface energy and preferential formation. Othercrystallographic features are identified based on their relative position to these facets ; i.e.Au (110) is located between two Au (111) facets, Au (311) is located between Au(100) andAu(111), Au(210) is located between Au (100) and Au (110). A side-by-side comparison of thebright field image to the fluorescence image of a DNA SAM is shown in Figure 4.8 with thesurface crystallographic regions of interest highlighted.The spherical nature of the bead electrode surface poses some challenges to the imagingprocess compared to performing measurements on a flat surface. The bead electrode wasaligned such that the regions of interest on the surface were within the depth of field. Withthe 5x objective, the minimum depth of field, is calculated through equation 4.1 where λ is thewavelength of light, η is the refractive index and NA is the numerical aperture of the objective.16The depth of field is 80 μm in the case of imaging point sources of light. In reality, the depth offield is about twice as thick (~150 μm) as the samples here are illuminated coherently.zmin =2λη(NA)2=2(0.5µm)1.33(0.13)2(4.1)When imaging the spherical electrode (with 1.5 mm diameter) where the bottom of thebead electrode and the surrounding surface are in focus, the depth of field includes a circularregion with a diameter of 0.9 mm. Thus, the majority of the electrode surface will be in focusat this magnification. The electrode was oriented in a way such that the stereographic triangleis situated near the bottom of the electrode and in the center of the field of view.Fluorescence images obtained were converted from their initial 16 bit format (integer) to32 bit format (float) to prevent loss of data through rounding in further calculations and toallow for negative numbers in further calculations. The images were then despeckled andGaussian blurred (2 px. radius) to reduce propagation of noise when performing mathematicaloperations during further analysis. A demonstration of this noise propagation reduction isshown in Appendix B. All fluorescence images in this thesis herein were background corrected734.2. CharacterizationFigure 4.7: Brightfield image of (a) a spherical single crystal bead electrode where (b) theAu(111) facets and (c) Au(100) facets are located. These are indicated by flat regions sur-rounded by terraces.Figure 4.8: a) Brightfield image of single crystal bead electrode with green ellipsoids indicatingregions of interest of different surface crystallographic features. b) The fluorescence image ofa DNA SAM assembled on the same electrode with two stereographic triangles containing thecrystallographic features labelled. Adapted from Leung and colleagues.168 Copyright (2018)with permission from Elsevier.744.2. CharacterizationCrystallographic Feature Range of pixels within each ROI (pix) Area Range (μm2){111} 4000-7000 12345-21604{100} 500-1000 1543-3086{110} 150-500 463-1543{311} 150-400 463-1235{210} 150-400 463-1235Table 4.1: Approximate size per region of interest for each crystallographic feature in numberof pixels and in μm2by subtracting the fluorescence image of the electrode without the DNA SAM. Additional dataprocessing such as measuring changes in fluorescence or normalization were done usingthe image calculator function in ImageJ which allowed subtraction and division of images.Unprocessed images are displayed here using a false colour lookup table for better contrast.Fluorescence intensity measurements listed in the work are the average intensity over mul-tiple pixels in a given ellipsoid region of interest (ROI). Green ellipses in Figure 4.8a highlightsuch ROIs from which fluorescence intensities were measured. The number of pixels de-pended on the size of the ROI. Selected crystallographic features and the approximate size ofeach ROI in number of pixels and in area are listed in Table 4.1. The fluorescence intensitiesin this thesis are shown in counts taken over a 3 second exposure time.4.2.3 Electrochemical DNA Coverage MeasurementsElectrochemical DNA coverage measurements were done on select DNA monolayers aftercharacterization using electrochemical fluorescence microscopy. Ru(NH3)6Cl3 (RuHex), a re-dox active ion, dissolved in the electrolyte is capable of electrostatically interacting with thenegatively charged phosphate DNA backbone.115,120 Cyclic voltammetry measurements andchronocoulometry measurements were used to detect RuHex ions. The parameters used aredescribed below.For cyclic voltammetry measurements, the DNA SAM was first allowed to equilibrate withthe electrolyte containing RuHex for 15 minutes where the RuHex in solution can interact withall the DNA in the SAM. The concentration of RuHex in the electrolyte was 2.5 μM (for lowDNA coverage) or 5 μM (for high DNA coverages) RuHex. The potential was cycled four timesbetween -0.45 V and 0.05 V vs. SCE at a fast scan rate (v = 0.5 V/s) and the DNA coverage is754.2. CharacterizationFigure 4.9: Cyclic voltammograms obtained from the first scan between -0.45V and 0.05 Vvs. SCE at scan rate 0.5 V/s for individual DNA SAMs on different Au single crystal beadelectrodes immersed in a) 2.5 μM and b) 5 μM of RuHex. Gray shading indicates the area ofthe cathodic peak which was used to calculate the DNA coverage. The DNA coverage of theDNA SAM in a) was low (1.07× 1012 DNA molecules/cm2) and in b) was high (4.43× 1012 DNAmolecules/cm2).obtained through the cathodic peak in the first scan located near -0.3 V/SCE.(Figure4.9)The charge (Q) due to reduction of RuHex at the surface can be calculated by dividing thecathodic peak area (Ap) by the scan rate (v) as in equation 4.2. The charge is proportional tothe amount of RuHex in a given area (A) on the surface (ΓRuHex, mol/cm2) through Faraday’sconstant (F). (Equation 4.3) The amount of RuHex is stoichiometrically related to the DNAcoverage (ΓDNA, DNA molecules/cm2) through equation 4.4 where NA is Avogadro’s number.Every RuHex ion, which has an oxidation state (z) of 3, electrostatically interacts with 3 DNAbasepairs, with each DNA strand at the surface containing 30 basepairs (m).37,115Q =APv(4.2)ΓRuHex =QFA(4.3)ΓDNA = ΓRuHex × zmNA (4.4)For chronocoulometry measurements, measurements were done first in the electrolytewithout RuHex and then in the electrolyte containing 50 μM RuHex. The DNA SAM was764.2. CharacterizationFigure 4.10: For chronocoulometric measurements,(a) a potential pulse from 0V to -0.45 V isapplied, (b) the current measured when applying -0.45 V is integrated (grey) to give the (c)charge and plotted against square root time. A y intercept is extrapolated from the charge dueto diffuse ions (- - -) to find the capacitative charge and charge from reduction of RuHex at thesurface.needed to equilibrate in the electrolyte containing 50 μM RuHex for 15 minutes prior to themeasurement. Pulses of 2 second duration were applied at 0 V/SCE then -0.45 V/SCE fol-lowed by application of 0.1 V/SCE for 0.01 s.(Figure4.10a) The current measured when apply-ing -0.45V/SCE (at 2 seconds) was integrated (Figure 4.10b) to give charge and plotted againstthe square root of the measurement time from 2 seconds. From the diffusion contribution tothe charge, a y intercept is extrapolated which indicates the total charge from capacitance andfaradaic reactions at the surface.(Figure 4.10c) The charge (from the y intercept) made in noRuHex was subtracted from the measurement made in the 50 μM RuHex in solution to givethe charge (Q) due to reduction of RuHex at the surface.1204.2.4 Correlating Fluorescence Microscopy Measurements to ElectrochemicalDNA Coverage MeasurementsThe fluorescence intensity measured during iSEFMI is related to the local surface coverage ofDNA at that specific region. It is assumed that the presence of organic species in the alka-nethiol and DNA SAM does not affect the fluorescence intensity. The intensity is only affectedby the quenching of the metal surface and the number of DNA molecules at the surface. To pro-vide a more useful metric to estimate the DNA coverage, a correlation between the local DNAcoverage and the fluorescence intensity measurements is required. This was accomplishedby using DNA SAMs that appeared to have an approximately homogeneous surface coverage774.2. CharacterizationFigure 4.11: (a) Fluorescence image of an example (relatively) homogeneous DNA SAM(made with immersion for 30 min in MCH, 1 hour in DNA at Edep =0.4 V/SCE) on a singlecrystal Au bead electrode. The region of interest, where the average fluorescence intensityof the entire electrode was measured, is shown as a black polygon. Select crystallographicfeatures are labeled to indicate the symmetry of the entire electrode with respect to the regionof interest. (b) The relation between this average fluorescence intensity and DNA coveragemeasured with electrochemical DNA coverage measurements in the presence of RuHex isshown. Reprinted with permission from Leung and colleagues.169 Copyright (2019) AmericanChemical Society.from iSEFMI images, and measuring the average DNA coverages using electrochemical mea-surements in a RuHex solution. These measurements provided the average DNA coveragefor the entire electrode, which was then correlated to the averaged fluorescence intensity. Dueto the difficulty of measuring the fluorescence intensity for the whole spherical single crystalbead electrode, the fluorescence intensity was taken in a region of interest which covered twostereographic triangles as shown in Figure 4.11a. Given the symmetric nature of the singlecrystal bead electrode, this fluorescence intensity was assumed to be the same on all otherstereographic triangles and therefore the entire electrode. The correlation of fluorescencemeasurement with DNA coverage is shown in Figure 4.11b. A linear correlation is observedat lower coverages (≤ 5× 1012 DNA molecules/cm2). However, the fluorescence intensity ap-proaches and slowly increases for coverages above 5×1012 DNA molecules/cm2 showing thathigher fluorescence intensities are less sensitive to changes in DNA coverage.78Chapter 5Formation of DNA SAMs viaPotential-Assisted Thiol-Exchange5.1 IntroductionThe conventional method for preparing DNA SAMs involves immersion of the gold surface ina solution of alkanethiol modified DNA followed by immersion in a MCH solution, which resultsin a mixed DNA/MCH SAM. The MCH is used to remove non-specifically adsorbed DNA (e.g.,those not adsorbed via thiol moiety)89,91 but also exchanges with specifically adsorbed DNAon the surface. This procedure will be referred to as DNA adsorption onto clean gold. An al-ternative approach where the immersion steps are reversed, dubbed the DNA thiol-exchangemethod, involves the alkanethiol-modified DNA in solution exchanging with an adsorbed MCHSAM. The thiol-exchange method results in a low coverage DNA SAM with less DNA aggre-gates and less non-specifically adsorbed DNA.99,103While these are clear advantages with the DNA thiol-exchange method, the resulting DNASAMs tend to be heterogeneous (i.e. have regions of different DNA coverages across thesurface), since thiol-exchange occurs first at defect sites and domain boundaries.56,57 Addi-tionally, thiol-exchange on sputtered gold surfaces was shown to occur at a slower rate withSAMs that are tightly packed.56,57 As a result, the surface crystallography, which controls thenumber of defects and packing of the initial alkanethiol SAM, will also have an influence onthiol-exchange.19 This has been previously shown by Yu and colleagues where the resultingDNA SAMs made via thiol-exchange on single crystal bead electrodes. Based on fluorescenceintensity measurements, DNA coverages were approximately 50% higher on the (100) com-pared to the (111) as a result of the higher packed MCH SAM on the (111). The high index795.1. Introductionplane (210) had a low surface coverage which contrasted the behaviour of another high indexplane (311). Despite both being high index planes, the opposing behaviour suggested that therelation between the surface crystallography and the packing of the initial MCH SAM was notas simple as a correlation of the atomically roughness with packing density.19 To the knowl-edge of the thesis author, potential-assisted thiol exchange, where a potential is applied to theinitial alkanethiol SAM (without desorbing the alkanethiol SAM) in the presence of a secondalkanethiol, has not be explored in literature.To understand the effect of an applied potential to the thiol-exchange process (i.e. potential-assisted thiol-exchange), change in coverage as a result of the applied potential to the initialMCH layer should be explored. Previous work has shown that applying specific potentials alteralkanethiol SAMs by creation of defects via reorganization of the SAM.170–173 At neutral pH, theSAMs are generally stable between −0.6 V and 0.6 V (vs SCE). Beyond this range, reductiveand oxidative desorption occurs.14,174,175 Within this range, based on EIS measurements inbuffered electrolyte, the ion permeability of an alkanethiol SAM was reported to be small andconstant in the positive potential range but an increase in permeability, due to reorganization ofthe layer, was observed at potentials more negative than a critical potential (which ranged from+0.2V to −0.3 V depending on chain length and headgroup of the alkanethiol).170,171,173 Above0.5 V, an increase in ion-permittivity has been detected with EIS for an alkanethiol SAM in0.1 M KClO4.176 No reorganization of the SAM has been detected with IR spectroscopy (PM-IRRAS),177 though it may not be sensitive enough to detect subtle increases in the numberof defects in the SAM. However, STM measurements of alkanethiol SAMs on Au(111) andAu(100) have shown structural changes when stepping the potential between -0.2 V to 0.55 Vin sulphuric acid.178–180While only scanning probe microscopy measurements are able to display defect creationor reorganization of the SAM under potential control, measuring the extent of thiol exchangecould provide another sensitive way of monitoring the increased number of defects as a resultof applied potential. The creation of defects from an applied potential as a function of theunderlying surface crystallographies is unexplored. As commercial substrates have polycrys-talline surfaces, being able to create identical SAMs on all surface crystallographies is impor-tant. This has already been proven difficult to do on single crystal bead electrodes,19 but it805.2. Experimentalis hoped that potential-assisted deposition can provide a means to create defects consistentlyon all surface crystallographies. The pzc depends on the surface crystallography,19,21,71,73therefore at a given potential, the number of defects created on each surface may not be thesame. As a result, potential-assisted deposition of DNA by thiol-exchange would likely resultin a DNA SAM that differs from that shown by Yu and colleagues.19In this chapter, DNA SAMs will be made through DNA thiol-exchange during the applica-tion of a constant potential (Edep) to a preformed MCH SAM on a single crystal bead elec-trode. The resulting modification will be evaluated using in-situ electrochemical fluorescencemicroscopy imaging (iSEFMI). This approach also provides a sensitive way of observing struc-tural changes made to the MCH SAM as a function of surface crystallography as a result ofthe applied potential. The surface coverage and local environment around the DNA will beexamined for the DNA SAMs made with thiol-exchange on five surface crystallographies asa function of the applied potential during deposition. Finally, DNA SAMs made on clean goldwhile applying a constant potential during DNA adsorption, as was previously done by Peter-son and colleagues,15 will be prepared. These SAMs will be characterized using iSEFMI anddirectly compared with DNA SAMs made with thiol-exchange at a constant potential.5.2 ExperimentalPotential-assisted thiol-exchange involved initially passivating the Au surface with MCH fol-lowed by DNA deposition while a potential (Edep) was applied (Figure 5.1a). After performingthe cleaning procedures described in section 4.1.1, Au electrodes were immersed 1 mM MCHdiluted in MeOH for 30 minutes. They were then rinsed and stored in MeOH until modificationwith DNA. Before DNA modification, the MCH coated electrode was rinsed with Millipore waterbefore immersing into immobilization buffer (IB) (80 μL) containing 1 μM of the fluorescentlylabelled alkanethiol-modified DNA. The IB used for DNA deposition in this chapter consistedof a 10 mM TRIS pH 7.5 buffer containing 100 mM NaCl (≥99.5% BioXtra, Sigma Aldrich)and 500 mM MgCl2 (>99% Sigma Aldrich). A saturated calomel reference electrode (SCE)was placed into the eppendorf via a salt bridge made of a glass pipette filled with 0.1 M NaClconnected to a teflon stopcock (Figure 5.1b).815.2. ExperimentalFigure 5.1: a) Schematic of the formation of a DNA SAMs using potential-assisted DNA thiol-exchange. b) Depiction of the 2-electrode apparatus used for potential-assisted DNA deposi-tion. Adapted from Leung and colleagues.168 Copyright (2018) with permission from Elsevier.The gold electrode was connected as the working electrode and the SCE was connectedas the reference electrode to an Autolab potentiostat. Controlled through the Nova softwareand via the potentiostat, a potential was applied across the working electrode and the refer-ence electrode. Initially the open circuit potential (OCP) is measured. The potential recordedfrom the OCP measurement was applied while an electrochemical impedance spectroscopy(EIS) measurement was measured using a 10 mV peak to peak sinusoidal perturbation withfrequencies from 1000 Hz to 1 Hz (Figure 5.2a). EIS measurements between 10 Hz and 1000Hz were fit to a RC circuit so as to characterize the initial MCH SAM based on the capacitancegenerated from the fit. For thiol-exchange with an applied potential, the deposition potential(Edep) was then applied whilst EIS measurements were made repeatedly (from 1000 Hz to1 Hz) approximately every minute over the immersion time (5 minutes to 60 minutes). Forthiol-exchange without potential control, the OCP was measured for a minute and then an EISmeasurement was obtained at this OCP. OCP measurements and EIS measurements wererepeatedly alternated for the remainder of the immersion time (Figure 5.2b). In both cases,after deposition, the gold electrode was disconnected from the potentiostat, removed from theDNA solution and then rinsed with Millipore water. The MCH/DNA modified Au electode was825.3. Results and DiscussionFigure 5.2: Potential profile over time during application of (a)Edep where the OCP is measuredfollowed by an EIS measurement then Edep is applied while taking multiple EIS measurements.Edep=+0.4V/SCE is shown here as an example. (b) For depositions at OCP, the OCP is mea-sured for some time and then a EIS measurement is taken at the last measured OCP.stored in IB overnight before characterization using iSEFMI.Potential-assisted DNA adsorption on clean Au followed a procedure similar to potential-assisted thiol-exchange. The cleaned Au electrode, placed in a 1 μM DNA solution in theIB, was the working electrode in the two-electrode apparatus described above. The OCPwas measured and then applied during an EIS measurement using the same parametersas above. Following this, Edep was applied for the length of the deposition (5 minutes) withEIS measurements repeatedly made. Depositions done without an applied potential involvedalternating measuring the OCP and measuring EIS at the measured OCP for 5 minutes. TheDNA modified electrodes were then rinsed with millipore water then immersed for 90 minutesin a 1 mM MCH solution diluted in the IB. The DNA/MCH SAM was rinsed with Millipore waterand stored in IB overnight prior to characterization.5.3 Results and Discussion5.3.1 Characterizing the MCH SAMThe thiol-exchange process investigated in this work relies on reproducible preparation of theMCH SAM (immersed for only 30 min) on the numerous gold bead electrodes used. The MCHSAMs prepared were characterized using differential capacitance measurements done with835.3. Results and Discussiona 5 mV 200 Hz potential perturbation. The potential was scanned from 0 V to either posi-tive or negative potentials (Figure 5.3) in an electrolyte identical to the immobilization bufferused in the thiol-exchange deposition procedure. This was to investigate the defects createdby making potential or negative excursions from 0 V and without desorbing the SAM prior tocharacterization. The capacitance was also measured with EIS in the same two electrodeconfiguration used for potential assisted DNA deposition with the capacitance calculated as-suming a simple RC circuit. The results are consistent with scanning differential capacitancemeasurements in the same electrolyte in a three electrode configuration. The capacitanceat 0 V/SCE for the MCH SAM modified gold bead electrodes measured in IB was 3.1 μF/cm2(±0.1 μF/cm2, N = 3). For comparison, the capacitance of a MCH SAM prepared for 24 h was2.76 μF/cm2, indicating the MCH SAM made with 30 minute immersion time was close to butnot at a maximum coverage, thereby facilitating the exchange process. The capacitance of aMCH SAM modified electrode slightly increased at potentials positive or negative of 0 V/SCE(Figure 5.3). This may be due to an increase in the more permeable regions in the SAM170,171suggesting that potential may influence the thiol-exchange process enabling DNA insertioninto the MCH SAM. As most surface modification processes are performed without potentialcontrol, the thiol-exchange at open circuit potential (OCP) is described first. Potentials chosenfor the controlled deposition were outside the OCP range where the capacitance is slightlyabove the minimum (Edep = −0.55, −0.45, +0.40, +0.50 V).5.3.2 Thiol Exchange of MCH layers with DNA at OCPModification of a MCH-coated single crystal gold bead electrode by DNA via thiol exchangewas performed at OCP that duplicates typical SAM deposition procedures. The initial OCP ofthe MCH SAM modified gold bead immersed in the DNA deposition solution was measured for40 samples. The average OCP was −8 mV with a range of values that span 350 mV (Figure5.4). This variation may be a result of slight differences in the electrolyte composition or in thequality of the MCH SAM. The OCP measured in the DNA deposition solution did not changesignificantly from the initial value even after one hour (Figure 5.5). These measurements wereperformed in the presence of O2 suggesting that the potential may be controlled by the redox845.3. Results and DiscussionFigure 5.3: Capacitance measured for a MCH SAM coated gold bead electrode in immobiliza-tion buffer. The line shows the capacitance measured from scanning the potential (with 5 mVrms 200 Hz perturbation ) from 0 V to 0.6 V or 0 to -0.6 V done using a three electrode setup.The symbols show measurements in the 2-electrode configuration used for the DNA deposi-tion at discrete potentials starting at 0 V stepping to 0.5 V and starting at 0 V stepping to −0.6 V.Adapted from Leung and colleagues.168 Copyright (2018) with permission from Elsevier.species in the deposition solution (e.g. reduction of O2). The OCPs adopted during depositionwere within the constant capacitance region in Figure 5.3 suggesting that the MCH layers weremostly well formed and stable.Any thiol exchange would be most active in regions on the surface where non-ideal SAMassembly occurred or where defects are located. SAMs made via thiol-exchange at OCP werecompared with those created using an applied potential to the MCH SAM during DNA depo-sition. These mixed MCH/DNA SAMs were later evaluated using iSEFMI after deposition wascomplete, but monitoring the thiol-exchange process (either with or without applied potential)in-situ can be valuable and was done using EIS.5.3.3 Capacitance measurements during DNA thiol exchange with MCH SAMBefore applying the deposition potential, the OCP was measured and the capacitance (atOCP) was characterized with EIS . Afterwards, the deposition potential was applied duringwhich the capacitance was measured during the hour-long deposition as shown in Figure 5.6.Since the surface areas of the electrodes used are not the same and using the assumption ofa consistent MCH SAM, the capacitance at OCP was set at 3.1 μF/cm2. Depositions at OCP855.3. Results and DiscussionFigure 5.4: Distribution of the initial open circuit potential measured for MCH SAMs upon im-mersion in the immobilization buffer containing DNA. Adapted from Leung and colleagues.168Copyright (2018) with permission from Elsevier.Figure 5.5: The open circuit potential (OCP) of MCH SAM covered gold bead electrodes whenimmersed in the DNA deposition solution for various deposition times (5 to 60 min). The OCPdoes not converge to a constant value. Adapted from Leung and colleagues.168 Copyright(2018) with permission from Elsevier.865.3. Results and Discussion0. (µF/cm2 )-0.55 V -0.45 V0. 0  10  20  30  40  50  60C (µF/cm2 )Time (min)OCP 0  10  20  30  40  50  60Time (min)0.40 VFigure 5.6: Capacitance measurements from fitting of data from impedance spectroscopy col-lected during DNA deposition at various Edep. Measurements were normalized by fixing theinitial capacitance measurement made at OCP (3.1μF/cm2). Colours correspond to replicateson different electrodes with errorbars corresponding to the estimated error of the capacitancedetermined based on the impedance data fit. Adapted from Leung and colleagues.168 Copy-right (2018) with permission from Elsevier.resulted in a gradual increase in capacitance which was not consistent from sample to sample.This is likely due to wetting of the stem of the bead electrode, increasing the surface area andtherefore capacitance over time (as reported in references69,172 for mercapto-undecanol). Incontrast, thiol-exchange occurring at Edep = +0.40 V/SCE resulted in an immediate increase incapacitance (∼20%) followed by a gradual decrease to 3.5 μF/cm2. The sudden increase maybe due to the creation of defects in the MCH layer while the steady decrease can be explainedby the reorganization of the layer as alkanethiol-modified DNA populates or exchanges withthese defects. These changes were quite consistent across different electrodes.In contrast, performing the DNA thiol-exchange at negative applied potentials (Edep= −0.45 V/SCEand −0.55 V/SCE) resulted in a gradual increase in the capacitance. This is a result of the ap-plied potential being near the reductive desorption potential for MCH (E<−0.6 V/SCE) and/ormay be due to wetting which is known to be more prominent at negative potentials.172,181 Ca-pacitance is sensitive to changes in SAMs but only reports the average behaviour and for these875.3. Results and Discussionelectrodes, cannot reveal any surface crystallography-specific data. In all cases, the increasein capacitance may indicate the presence of DNA in the SAM. Fluorescence microscopy willenable characterization of the modified single crystal Au bead electrodes providing an estimateof the extent and homogeneity of DNA thiol-exchange.5.3.4 Fluorescence Characterization of MCH/DNA layersThe extent of DNA thiol-exchange into the MCH SAM at either OCP or at Edep= +0.40 V/SCEwas characterized using iSEFMI. Edep=+0.4V/SCE was chosen due to the observed increasein defects created in the MCH SAM (shown previously with capacitance measurements) butwithout desorption of the MCH SAM. This approach can assess the DNA distribution on theelectrode surface and on the various crystallographic regions for both MCH/DNA layers madeat OCP and prepared using an applied potential. The fluorophore label on the DNA and theresulting fluorescence intensity can be used to estimate the extent of DNA thiol-exchange.Fluorescence images of the MCH/DNA SAMs prepared by thiol-exchange at OCP and atEdep= +0.40 V/SCE as a function of the immersion time are compared in Figure 5.7. Theimages of the single crystal bead surfaces were indexed using methods described in Yu andcolleagues19 which enabled analysis of the average fluorescence intensity as a function ofsurface crystallography and immersion time (Figure 5.9 for low index surface crystallographiesand Figure 5.10 for high index surface crystallographies).It should be noted that the thiol-exchange performed at OCP did involve a period of timewhere the OCP was applied during EIS measurements. This periodic application of the OCPdid not result in a thiol-exchanged surface that was different than the ones prepared withoutthis EIS measurement (Figure 5.8).The thiol-exchange process at Edep= +0.40 V/SCE shows that the low index planes (111),(100), (110) have a significantly brighter fluorescence and therefore higher surface concen-tration of DNA as compared to those prepared at OCP (Figure 5.9). The rate of DNA thiol-exchange was also faster at +0.40 V/SCE as compared to OCP. Since images of the bead sur-face contain two or more stereographic triangles, a comparison of the deposition on differentregions with similar crystallography is possible. It is important to note that the spherical nature885.3. Results and DiscussionFigure 5.7: Fluorescence images taken of MCH/DNA layers prepared a) at OCP (no appliedpotential) and b) at Edep= +0.40 V/SCE. Images from left to right correspond to increasingtime in the deposition solution. Each image is from a different electrode resulting in a differentorientation. The stereographic triangle and crystallographic regions analyzed are shown onthe images. All images are false coloured to represent intensity. Adapted from Leung andcolleagues.168 Copyright (2018) with permission from Elsevier.Figure 5.8: Comparison of the fluorescence images (left) and fluorescence intensity (right)measured from selected regions on the gold bead electrode after 60 min thiol-exchange pro-cess performed at OCP with the application of the OCP during EIS measurements to that withonly monitoring the OCP. Each bar corresponds to a single measurement with errorbars cor-respond to the standard deviation of measured fluorescence intensities on a region of interestwith a given surface crystallography. Adapted from Leung and colleagues.168 Copyright (2018)with permission from Elsevier.895.3. Results and DiscussionFigure 5.9: Fluorescence intensity, which correlates to the DNA coverage, measured acrossdifferent DNA immersion times, measured on the low-index regions (111), (100), and (110), forMCH/DNA layers made at OCP (open points) and layers made at Edep= +0.40 V vs SCE (filledpoints). Individual data points correspond to individual replicates with errorbars indicatingstandard deviation of fluorescence intensity measurement within the region of interest mea-sured. N=2-3 for each condition. Adapted from Leung and colleagues.168 Copyright (2018)with permission from Elsevier.Figure 5.10: Fluorescence intensities measured from MCH/DNA on gold bead electrodes pre-pared via thiol-exchange at Edep= +0.40 V vs. SCE (filled points) and at OCP (open points)for the high index regions (311) and (210). Individual data points correspond to individualreplicates with errorbars indicating standard deviation of fluorescence intensity measurementwithin the region of interest measured. N=2-3 for each condition. Adapted from Leung andcolleagues.168 Copyright (2018) with permission from Elsevier.905.3. Results and Discussionof the beads presents some challenges to measuring consistent intensities. For example, oneof the (111) facets appears to be more intense than the others. This is a consequence of differ-ences in the illumination/collection efficiencies due to the curvature of the sample. Thereforethe stereographic triangle of interest was positioned so that it was in the focal plane and theanalysis was done on these regions. A more uniform deposition was achieved at +0.40 V/SCEat all the times investigated, whereas deposition at OCP was significantly more variable. Therange of intensities on the (111) facets was large for the 1 h deposition at OCP (σv= 20%) ascompared to (σv= 5%) for Edep= +0.40 V/SCE deposition.Electrochemical DNA coverage measurements with RuHex revealed a DNA coverage of9.5(±0.8) × 1012 DNA molecules/cm2 for electrode surfaces prepared at +0.40 V/SCE, whichis ~10x larger than those layers prepared at OCP (8.8(±0.4)×1011 molecules/cm2). These are5 and 50 times smaller than the maximum DNA coverage theoretically possible15,37 but matchthe range where hybridisation efficiency is high.15 While the average surface concentration isin the optimum range for sensing, knowing the distribution of DNA on the surface is importantsince the local surface coverage is high in some regions and low in others. This would resultin a significant difference in the DNA hybridisation efficiency on the same electrode surfaceand therefore influence the ability of the DNA sensor to detect a complimentary strand. Thisis exemplified in the DNA coverages on the (311) and (210) surfaces (Figure 5.10). Both areturning points in their respective crystallographic zone axes and therefore represent steppedsurfaces with the (210) known to have the highest surface energy and greatest atomic rough-ness among all gold surface crystallographies.19 The (311) surface is exchanged efficiently,nearly independent of the application of potential, while the (210) surface is devoid of DNA.A significant variation in the extent of DNA thiol-exchange on the five regions analyzed isobserved as a function of the immersion time for the applied potential (+0.40 V/SCE) and lessso at OCP. The reproducibility of the fluorescence intensity is less than desired, as the variationin fluorescence intensity indicate at most a 5 × 1012 DNA molecules/cm2 range in coverage.This shows how difficult simple deposition procedures are to reproduce when performing anal-ysis of surface regions rather than just the average behaviour. Interestingly, the three low indexplanes do not behave similarly, with the (111) surface becoming modified more quickly thaneither of the (100) or (110) surfaces. After 60 minutes, they reach the same intensity value.915.3. Results and DiscussionThis result suggests that the extent of thiol-exchange is dependent on the characteristics ofthe MCH SAM and how the defect density or permeability changes with positive potential. Inessence, the thiol-exchange does not involve a significant change to the gold - thiol interac-tion since both species are interacting with the surface via a C6 alkanethiol moiety. But anincreased DNA surface coverage is favored since the alkanethiol modified DNA is the onlysurface active species in solution. The prevalence of DNA thiol-exchange on specific crystal-lographic regions may be a consequence of more defects in the initial MCH SAM, or due to afavorable interadsorbate interaction between DNA strands as a result of the high [Mg2+] in theIB used, that encourages thiol-exchange at positions where the surface is modified with DNA.The (311) surface is quickly populated with DNA followed by the (111) facet (in addition to theregion around the (111) which can be considered large (111) terraces with large steps182).These surfaces can be considered as mainly (111) in nature; (311) can be considered as2(111)x100 in addition to 2(100)x111. The presence of (111) terraces along with steps maymaximize interadsorbate interactions between the DNA and/or the initial MCH layer is morelabile on these surfaces, resulting in a high DNA coverage. This would be mediated by thehigh ionic strength electrolyte providing significant shielding or even ionic salt bridging.183 Thesame trend was observed for OCP thiol-exchange, although lower in intensity and with lowerDNA coverage. Interestingly, the (210) (2(100)x110) surface contains little thiol-exchangedDNA for both deposition conditions suggesting a low defect density in the MCH SAM. The con-trast between both (311) and (210) crystallography, despite both being atomic rough surfaces,is noteworthy and shows that the thiol exchange behaviour is strongly influenced by the typeof terrace and step which dictates the strength of MCH adsorption, decreasing the rate of DNAadsorption or thiol exchange.195.3.5 Thiol-exchange at Various PotentialsThe thiol-exchange was performed for one hour at a more positive potential (Edep= +0.50V/SCE) and characterized using fluorescence microscopy (Figure 5.11a) and the fluorescenceintensities measured across the different crystallographic regions (Figure 5.12). The fluores-cence image reveals DNA thiol-exchange on every region of the electrode surface, includ-925.3. Results and DiscussionFigure 5.11: Fluorescence images taken of MCH/DNA layers made with at 1 h DNA immersiontimes at (a)Edep= +0.50 V, (b)Edep= -0.45 V and (c)Edep= -0.55 V. Image Exposure time was3s with EM gain 50 and falsely coloured to emphasize fluorescence intensities.Adapted fromLeung and colleagues.168 Copyright (2018) with permission from Elsevier.ing (210). Based on the relative fluorescence intensities, DNA deposited at a higher surfacecoverage in the regions surrounding the (111) facet and on the (100) facet. The (110) re-gion observes a decrease in fluorescence intensity compared to the (110) when Edep= +0.40V/SCE. This is likely due to the nature of defect creation as a result of the applied potentialand the DNA adsorption process. As previously demonstrated through differential capacitancemeasurements of an MCH layer, the number of defects increase with more positive appliedpotentials as shown by the increase in capacitance with more positive potentials in Figure 5.3.During thiol-exchange, when a Edep= +0.40 V/SCE is applied, defects are created in the non-fluorescent MCH SAM, creating a new site for the fluorescently labeled DNA to adsorb. Thisresults in a net increase in fluorescence intensity. With a higher amount of defects at Edep=+0.50 V/SCE, a simultaneous destabilization of the newly adsorbed DNA can occur, resultingin a fluorescence intensity that is lower than seen at Edep= +0.40 V/SCE.The thiol-exchange process is observed to occur more homogeneously over all surfacecrystallographies, suggesting that the creation of defects with positive potential becomes moreuniform at Edep= +0.50 V/SCE as compared to +0.40 V/SCE. Oxidative desorption of alka-nethiol SAMs begins at +0.6 V/SCE in basic electrolyte72 and could be the underlying mech-anism for the increase in defects across the surface. Specific adsorption of chloride from theimmobilization buffer is also a likely source of defects at positive potentials, shifting the onset935.3. Results and DiscussionFigure 5.12: Fluorescence intensities measured from various crystallographic regions for thiol-exchanged MCH/DNA SAMs prepared for 1 h at different applied (or OCP) potentials as rawintensities. Errorbars correspond to standard deviation across replicate samples (N=3 to 6 perEdep). Adapted from Leung and colleagues.168 Copyright (2018) with permission from Elsevier.of oxidative desorption to less positive potentials. This will be explored further in Chapter 7. Atnegative potentials, reductive desorption via a defect driven growth of the holes in the SAM184begins at −0.6 V/SCE and provides an opportunity to compare the thiol-exchange process atnegative potentials with thiol-exchange at positive potentials.The capacitance of the MCH SAM (Figure 5.4) was shown to increase when applying neg-ative or positive potentials, suggesting similar defect formation. The potential-assisted thiol-exchange of DNA with an MCH SAM at Edep = −0.45 V/SCE and −0.55 V/SCE for one hourwas studied. The fluorescence images of the resulting thiol-exchanged surfaces are shown inFigure 5.11b and c. The largest DNA surface coverage is observed for the (311) and (111)regions in addition to around the (111) facet. The (100), (110) and (210) surface regions havelittle DNA at Edep = −0.45 V/SCE, but increases with more negative potential. Reductive des-orption of alkanethiol SAMs starts with the creation of defects184–186 and then proceeds fromthe (111) surface at the least negative potential19,71 with desorption from the (100) surfacerequiring more negative potentials. The defect creation disrupts the SAM and enables DNAto thiol-exchange or fill in the unoccupied regions. Interestingly, the DNA coverage does notchange when performed at Edep = −0.55 V/SCE, suggesting more than defect formation is im-portant, with reductive desorption becoming a competing process at these negative potentials.945.3. Results and DiscussionAs mentioned previously, the thiol modified DNA is the only thiol present in solution whichensures an increase in the DNA coverage. It is interesting that the fluorescence intensity (cov-erage) of the (111) facets are similar to that at positive potentials even with the small amount ofreductive desorption. The interadsorbate interaction between DNA strands may stabilize theSAM especially in the IB used (high [Mg2+]),187 thereby keeping the DNA specifically adsorbedand stable at these negative potentials. We do not observe non-specifically adsorbed DNAleaving the surface during the in-situ fluorescence measurements clearly indicating these arechemisorbed DNA molecules.140 This result demonstrates that for a negative applied potential,in high ionic strength electrolyte, the DNA is not electrostatically repelled from the surface andthiol-exchanges with defects in the layer. The balance between the thiol-exchange and reduc-tive desorption processes is important and these results show the influence of the underlyingsubstrate but further investigation is needed.The (311) region again has the highest fluorescence intensity and is therefore highly pop-ulated with DNA, even at these negative potentials suggesting that the thiol-exchange at the(311) region is effectively independent of the applied potential. This is further supported by themore negative reductive desorption potentials of the (311) surface compared to the the (111)region,19 which suggests that the presence of defects in the (311) allows facile thiol-exchangeor a DNA SAM organization that is particularly stable for the (311) region.A comparison of the fluorescence intensities from surfaces after one hour of thiol-exchangeat OCP or at the other applied potentials is shown in Figure 5.12 for the (111), (100), (110),(311), and (210) regions. Thiol exchange on the (111) facet is enhanced at both positive andnegative potentials when compared to the OCP conditions. The low index planes (111), (100),(110) contain more DNA at the positive potentials as compared to the negative potentials,reflecting a difference in the defects formed and the thiol-exchange process as a function ofpotential. The (210) region has the lowest DNA coverage for all conditions, increasing asthe potential is made either positive or negative. The distinct difference in the results fromnegative and positive deposition potentials may be due to the composition of the electrolyteused in these thiol-exchange processes and will be discussed in Chapter 7.955.3. Results and Discussion5.3.6 Assessing the Local Environment of the DNA SAMs made withPotential-Assisted Thiol-ExchangeAn important characteristic for nucleic acid sensors is the local environment of the tetheredDNA which ensures the complementary target strand can efficiently hybridize.15 The localpacking density of DNA can be estimated by measuring the ease of reorientation of the ad-sorbed DNA when driven by modulating the charge density on the gold surface.116,140,156,157Fluorescence images after DNA thiol-exchange were measured at −0.40 V/SCE where theDNA is electrostatically repelled from the surface making the terminal fluorophore furthestaway from the surface resulting in a maximum in fluorescence intensity without desorbing theSAM (Fmax). Another image was taken at +0.35 V/SCE, where the DNA is attracted to thesurface and terminal fluorophore is brought closer to the surface resulting in a minimum influorescence (Fmin). The change in fluorescence (Fmax − Fmin) is divided by the backgroundcorrected maximum fluorescence intensity (Fmax) to normalize for the variation in coverage(Figure 5.13). The modulation of fluorescence intensity is given in the following equation:Fmod =Fmax − FminFmax(5.1)If all the DNA is able to freely reorient, then Fmin = 0 and Fmod = 1, if the DNA is packedso tightly that it cannot reorient, then Fmin = Fmax and Fmod = 0. Analysis over the surfaceregions studied will provide an estimate of the local DNA packing, which may not be easilyestimated by simply measuring intensity. If the DNA on the surface was clustered togetherin structures below the optical resolution, there fluorescence intensity and the DNA coveragewould appear low, but this region would have Fmod = 0. Examples of this will be shown inChapter 7.The relation between the intermolecular interactions of the adsorbed DNA (i.e. when themovement of the DNA is hindered by neighbouring DNA) and the fluorescence modulationcan be modeled. In this model, the DNA behaves as a rigid rod with length lDNA during theiSEFMI measurement since a low ionic strength electrolyte was used (10 mM TRIS 10 mMKNO3).88 The effective diameter (deff ) of the DNA strand is a sum of the diameter of DNA withthe Debye length and increases with decreasing ionic strength.188 The model assumes that965.3. Results and DiscussionFigure 5.13: Example calculation of the fluorescence modulation of a particular region of in-terest on a DNA SAM. Images taken during iSEFMI of the (111) at -0.4V (Fmax) and +0.35 V(Fmin) are shown. Images showing the subsequent calculations for Fmax−Fmin and Fmod aredisplayed.the DNA SAM is arranged equidistant in a hexagonal arrangement. The distance between thetethered ends two adsorbed DNA (La, in nm) is shown in the following equation 5.2 with theDNA coverage (ΓDNA) in molecules/cm2.La =√1014(√32× ΓDNA)−1 (5.2)Two possible limiting cases for DNA reorientation are depicted in Figure 5.14. The max-imum angle of motion (φ) for a DNA is sterically limited and depends on separation from itslocal neighbors with consideration of the ionic strength dependent diameter of the DNA. In onecase (Figure 5.14a), the tethered DNA can move together as described by Li188, or two DNAstrands can reorient so as to encounter each other, the strands will precess at an angle withrespect to the surface, limited by the same motions of its neighbor (Figure 5.14b). Calculatingφ for a DNA of fixed lDNA and deff can be done using equations 5.3 and 5.4 for models aand b respectively. In both models when La is equivalent to deff , the DNA does not move (i.e.φ = 0◦).975.3. Results and Discussionfor deff<La<lDNA cos(φ) = deff/Lafor La > lDNA φ = 90◦(5.3)for deff < La < 2lDNA sin(φ) =(La2− deff2)lDNAfor La > 2lDNA φ = 90◦(5.4)In both models, φ decreases as the distance between the adjacent DNA (La) decreasesbut at a greater extent for model B. The maximum fluorescence modulation is calculated basedon the maximum angular motion of the DNA from both model a and model b. Since the flu-orescence intensity is proportional to the cube of the separation between the fluorophore (atthe distal end of the DNA) and the gold surface (Equation 5.5 and 5.6),16,75 the change inthe fluorescence intensity can be calculated using equation 5.7. An estimate of the changein fluorescence intensity due to DNA reorientation as a function of DNA coverage is shown inFigure 5.14c for model a (black) and model b (red). Using the relation between DNA cover-age and maximum fluorescence intensity measured at -0.4V/SCE (from 4.2.4), the calculatedfluorescence modulation and the maximum fluorescence intensity is shown in Figure 5.14d.Fmax  l3DNA (5.5)Fmin  (lDNA × cosφ)3 (5.6)Fmod =Fmax − FminFmax= 1− FminFmax(5.7)Fmod = 1− cos3 φ (5.8)The fluorescence intensity and the fluorescence modulation was measured for all DNASAMs prepared via thiol-exchange (at OCP or with an applied potential) in the regions of crys-tallographic interest. Fmod was determined and compared to fluorescence intensity in Figure5.15. The top x axis indicates the DNA coverage which was correlated to the fluorescence985.3. Results and DiscussionFigure 5.14: Models (a and b) of the extent of DNA angular mobility during potential drivenDNA reorientation with a cylindrical framework for the DNA (green) and a fluorophore at thedistal end (red circle). A modeled estimate of the fluorescence modulation for both types ofDNA reorientation as a function of c) DNA surface coverage or d) the estimated fluorescenceintensity from iSEFMI measurements.intensity based on measurements in section 4.2.4. A decrease in Fmod is observed for DNAcoverages greater than ~1012molecules/cm2 as inter-adsorbate interaction becomes impor-tant, similar to what was reported by Rant.118 The highest fluorescence intensity regions showno fluorescence modulation since the DNA coverage is close to the theoretical maximum (2to 3× 1013 DNA/cm2),15 where the DNA movement is restricted and unable to respond to thepotential modulation.The fluorescence modulation from all the surfaces studied after preparation via thiol-exchangefall on a universal curve and are generally independent of the deposition potential or the crys-tallographic region analyzed. This experimentally determined curve lies between the two mod-elled curves shown in Figure 5.14d. This is understandable as both models are possible andthe fluorescence modulation would be a weighted average of the presence of these two mod-els on the electrode surface. At higher coverages, the DNA will tend to move together as inmodel a, and at lower coverages, steric restrictions shown in model b are more likely to occur.The relationship between the fluorescence modulation and the maximum fluorescence in-tensity are shown separately in Figure 5.16 for each crystallographic region studied. In general,these results show that the thiol-exchange procedure creates mixed DNA/MCH SAMs that be-have as if the DNA exists in the same local environment on different parts of the bead electrodesurface; that is, for a DNA SAM on a given surface crystallography, the DNA is homogeneouslydistributed.With one hour deposition time, the potential-assisted thiol-exchange process is capable995.3. Results and Discussion 0.01 0.1 1 10 100102 103 1041012 1013 1014Fl. Mod. (%)Fl. Int. (counts)ΓDNA(molecules/cm2)(111)(100)(110)(311)(210)Figure 5.15: Analysis of the ability of the DNA SAM to undergo potential driven reorientation(resulting in a modulation of fluorescence intensity, Fmod) for a range of surface concentrations.Fluorescence Modulation plotted against fluorescence intensity (bottom x axis) and DNA cov-erage (top x axis) for all regions of interest measured, displaying negative correlation betweenDNA coverage and DNA local crowdedness (- - -) until a maximum coverage is attained withno detectable fluorescence modulation.The DNA coverage correlation with the maximum fluo-rescence intensity is based on measurements made in section 4.2.4.of producing DNA SAMs with DNA coverages in the 1012 DNA /cm2 range. The DNA SAMsat these coverages are optimal for use as hybridisation sensors given the sufficient spacingbetween the DNA at the surface.15 However, potential-assisted DNA thiol-exchange results inDNA SAMs with a wide range of coverages depending on the crystallographic nature of theelectrode. The typically used vapor deposited gold surfaces contain a variety of low index(typically (111)) and high index (grain boundaries) regions.26 The low index planes will beevenly modified with the DNA coverage depending on the applied potential and depositiontime. The high index planes may well contain no DNA like the (210), or so much DNA likethe (311) that it would not be active in the sensing process. This work outlines the importantparameters that can be modified so that the thiol-exchange procedure will be able to preparesurfaces with well defined coverage, evenly distributed across all surface crystallographies.1005.3. Results and Discussion 0.01 0.1 1 10 100102 103 1041012 1013 1014Fl. Mod. (%)Fl. Int. (counts)ΓDNA(molecules/cm2)(111)102 103 1041012 1013 1014Fl. Int. (counts)ΓDNA(molecules/cm2)(100)102 103 1041012 1013 1014Fl. Int. (counts)ΓDNA(molecules/cm2)(110) 0.01 0.1 1 10 100102 103 1041012 1013 1014Fl. Mod. (%)Fl. Int. (counts)ΓDNA(molecules/cm2)(311)102 103 1041012 1013 1014Fl. Int. (counts)ΓDNA(molecules/cm2)(210)Figure 5.16: The magnitude of the fluorescence modulation for DNA SAMs for the five crys-tallographic regions studied as a function of the fluorescence intensity measured. The Figureincludes DNA SAMs made via thiol-exchange at OCP (open points) and Edep=0.4 V/SCE (filledpoints). Adapted from Leung and colleagues.168 Copyright (2018) with permission from Else-vier.1015.3. Results and Discussion5.3.7 Fluorescence Characterization of DNA SAM Formation on Clean Au(DNA/MCH)Using potential-assisted deposition during the conventional DNA SAM formation method (i.e.DNA adsorption on clean Au) will be of interest to the biosensor community.15,147 Applying apotential to the gold surface has enabled a way to control the DNA coverage and also decreasethe manufacturing time for these DNA SAMs. In the conventional method, which was firstproposed by Herne and Tarlov,89 the gold surface is immersed in a DNA solution to form aninitial DNA SAM followed by thiol-exchange with MCH. Potential-assisted DNA SAM formationon clean gold involves applying a potential during the DNA adsorption step while the followingstep, involving the immersion of the layer in MCH, occurs at OCP.15 Previously, Murphy andcolleagues have directly compared DNA SAMs made with thiol-exchange to DNA SAMs madevia DNA adsorption on a clean gold surface at OCP.103 They found that compared to DNASAMs made with thiol-exchange, DNA SAMs made via adsorption onto a clean gold surfaceresulted in high DNA coverage and higher amounts of DNA aggregates and non-specificallyadsorbed DNA.In this section, DNA SAMs were made on single crystal bead electrodes via DNA adsorp-tion onto clean gold at OCP, at Edep = +0.40 V/SCE and Edep = +0.50 V/SCE for only 5 minutes.With this short DNA immersion time and without any applied potential (Figure 5.17a), the DNAcoverage is significantly higher in comparison to the DNA SAMs made via thiol-exchange atOCP with a 1 hour DNA immersion time (Figure 5.7a). The DNA coverage for one of theseSAMs was found to be 5.0 × 1012 DNA molecules/cm2, two times lower than the coverage ofthe SAM made using DNA thiol-exchange while applying Edep = +0.4V/SCE for 1 hour. Thehigh coverage can be attributed to the lack of MCH adsorbed to the surface, which would haveimpeded DNA adsorption. The absence of a MCH SAM also contributes to a more uniformdistribution across the many surface crystallographies in contrast to the DNA SAMs made withthe thiol-exchange method. The surfaces made using DNA adsorption onto clean gold displaya high coverage across most surface crystallographies, although there are subtle differencesin DNA coverage. Interestingly, the fluorescence intensity on the (311) surface is the small-est in contrast to the thiol-exchange method. This further confirms that the (311) surface is1025.3. Results and DiscussionFigure 5.17: Fluorescence Images of DNA SAMs made with 5 minute DNA adsorption atvarious potentials a)OCP b) Edep= +0.4 V c) b) Edep= +0.5 V. Adapted with permission fromLeung and colleagues.169 Copyright (2019) American Chemical Society.Figure 5.18: Bar graph showing average fluorescence intensities of DNA SAMs on certainsurface crystallographies. The errorbars shown in the bar graphs correspond to the larger ofeither the standard deviation of 2 replicates or the propagated measurement errors. Adaptedwith permission from Leung and colleagues.169 Copyright (2019) American Chemical Society.labile and thiol-exchange readily occurs there. In this case, a non-fluorescent thiol (MCH) isreplacing a fluorescently labeled DNA, resulting in the fluorescence intensity decrease.DNA SAM formation on clean gold during application of a constant potential (Edep = +0.4V/SCE) for 5 minutes reveals only a slight increase in DNA coverage (Figure 5.17b) com-pared to those prepared at OCP for 5 minutes. The measured coverage of 6.4 × 1012 DNAmolecules/cm2 was 20% greater than 5.0× 1012 DNA molecules/cm2 for the DNA SAM madeat OCP. Examining the fluorescence intensity on the five surface crystallographies of interest(Figure 5.18) showed that a noticeable increase in the fluorescence intensity was observedonly at the (210) surface.1035.4. ConclusionsApplying a more positive potential (Edep = +0.5 V/SCE) results in a slightly higher DNAcoverage (Figure 5.17c). However, when measuring the DNA coverage on individual surfacecrystallographies, a noticeable increase is seen on the (110) and (111), while the DNA cover-age does not increase on other surface crystallographies as seen in Figure 5.18. The smallincreases in the overall DNA coverage can be attributed to the defect creation as a result ofthe applied potential. During potential-assisted DNA adsorption on clean gold, the defectsformed resulted in destabilization of any weakly adsorbed DNA, preventing non-specificallyadsorbed DNA and providing additional sites for other DNA to adsorb.14 However, these in-creases are small due to the high coverage DNA SAMs already made with DNA adsorption onclean gold, minimizing the influence of of an applied potential. Compared to potential-assistedthiol-exchange, potential-assisted DNA SAM formation on clean gold does not seem able tocontrol the DNA coverage over a large range. This makes potential-assisted thiol-exchange amore suitable method for tailoring DNA SAMs with low to high coverages.5.4 ConclusionsPreparing a mixed fluorophore-labeled DNA and alkanethiol SAM via thiol-exchange usingan applied potential was studied using a single crystal bead electrode and fluorescence mi-croscopy to characterize the thiol-exchange on the gold (111), (100), (110), (210), (311) sur-faces at a variety of potentials. Compared to thiol-exchange that occurs at without an appliedpotential (or OCP which is 0 V/SCE on average), positive potentials (Edep= +0.4 and +0.5V/SCE) produced highly covered surfaces that were uniform across the low index planes. Forthe high index planes, the (210) surface had the lowest coverage while the (311) surface hadthe highest DNA coverage.A more positive potential produced a surface that was almost uniform in DNA thiol-exchangecoverage. The application of potential also increased the rate of DNA thiol-exchange producinghigh coverage in one hour compared to DNA thiol-exchange at OCP which created a surfacewith 10× lower coverage. Deposition at negative potentials was also studied. Thiol-exchangewas limited to the low index planes in addition to a high coverage (311) region. The formationof defects in the MCH SAM increased at these positive or negative potentials which enabled1045.4. ConclusionsDNA thiol-exchange, but the nature of the defects or competing desorption processes werehypothesized to control this process. The electrolyte did contain a high concentration of Cl–which is known to adsorb to gold surfaces and mobilize gold atoms.189–191 This may result increating defects in the MCH SAM at positive potentials and will be explored in Chapter 7.The local environment of the surface bound DNA prepared in this way was also studiedby measuring the extent of DNA reorientation (via modulation of the DNA with potential whilemeasuring the changes in fluorescence intensity) to gauge to the local surface concentration.All the surfaces and samples prepared fell on a common curve showing that for these sur-faces, the DNA arranged in relatively uniform manner at the nm scale. Surface concentrationsof 1012 molecules/cm2 were routinely prepared using electrodeposition for one hour. Potential-assisted DNA adsorption on clean gold on the other hand created high coverage layers (1013molecules/cm2) after only 5 minutes of deposition. It was found that potential-assisted DNA ad-sorption on clean gold for 5 minutes did not change the DNA coverage dramatically comparedto potential-assisted thiol-exchange for a deposition time of 5 minutes to 1 hour. It is clearthat the DNA coverage for different surface crystallographies can be affected using potential-assisted DNA thiol-exchange. By influencing the stability of the MCH layer present during theDNA thiol-exchange method with applied potential, it is possible to limit the amount of DNAadsorbed on a given surface crystallographic feature. Thus, a uniformly distributed DNA SAMcan be made on the surface of polycrystalline gold substrates used in DNA biosensors. Astrategically chosen Edep would enable modification of the entire polycrystalline surface, onlylow index/atomically flat surfaces, or even exclude higher index surfaces typically found atgrain boundaries. Having this level of control on the DNA coverage on the different surfacecrystallographies would allow for higher reproducibility in DNA sensor performance.105Chapter 6The Effect of a Square-WaveModulated Potential during DNADeposition6.1 IntroductionFacilitating the DNA SAMs formation in a fast, facile and controllable way is important for themass production and cheap manufacture of DNA biosensors. Recent literature has proposedthe application of a modulated potential during SAM formation. A square-wave modulatedpotential has been found to facilitate the adsorption of DNA on the gold electrode surface andreduce the time required for DNA self-assembly to 30 seconds to 5 minutes.147,192 However,a mechanism that can explain the increased DNA coverage in this short deposition time is stillnot well developed.Quan and colleagues found that switching the potential repetitively from positive to nega-tive values in a 50 mM triethyl ammonium acetate buffer for an hour resulted in higher DNAsurface coverage. This was measured with an electrochemical quartz crystal microbalancewith dissipation monitoring.192 The higher DNA coverage was speculated to be a result of theelectrostatic effects of both positive and negative potentials on the negatively charged phos-phate DNA backbone. Applying positive potentials during DNA deposition caused the DNA toadsorb via the thiol moiety and the DNA then became oriented perpendicular to the surfacedue to the negative potential. The authors believed that when the adsorbed DNA is orientedperpendicular to the surface, the number of available DNA binding sites increases as they wereoriginally blocked by the DNA oriented parallel to the surface.1066.1. IntroductionJambrec and colleagues applied a 50 Hz square wave with an amplitude of 0.7 V (from0.4V to -0.3 V/SCE) to a gold electrode that was immersed in a DNA solution with 10 mMphosphate buffer and 450 mM K2SO4. The higher ionic strength electrolyte would result inshielding of the negatively charged backbone, making it difficult for the DNA to be electro-statically actuated by the potential. Despite this, higher DNA coverages with fewer defects inthe SAM were obtained, suggesting that the mechanism proposed by Quan and colleagueswas not completely accurate.147 High coverage alkanelthiol SAMs were made using the samesquare-wave potential profile, implying that enhanced adsorption due to the square-wave po-tential does not require the adsorbate to be charged.193 The authors proposed that in the highionic strength buffer, the potential only affects the region near the electrode surface, and if thepotential limits used for the square wave were centered around the potential of zero charge(pzc) of the surface, there would be an exchange of cations and anions causing a stirring effectwhich transports more DNA to the surface thereby increasing its coverage.In the previous chapter, the role played by a constant applied potential (Edep) on thiol-exchange process and DNA adsorption on clean gold was examined. The DNA coveragewas found to be was strongly influenced by the applied potential, which was thought to createdefects in the SAM. The density of these defects varied depending on the potential and surfacecrystallography. Using potentials close to either reductive or oxidative cleavage of the Au-thiolbond created a large number of defects which was most effective for facilitating thiol-exchange.It has yet to be determined if these defects are also created by the square-wave modulatedpotential and is responsible for the enhanced DNA coverage seen by others.As both Quan and colleagues and Jambrec and colleagues utilised characterization tech-niques which report an average coverage for the polycrystalline gold surface, it is important toverify their findings using an imaging technique such as iSEFMI. This is to examine hetero-geneities in the DNA SAMs and also investigate the influence of surface crystallography, whichwould reduce the consistency of DNA SAMs made on polycrystalline surfaces. In this chapter,the effect of a modulated potential during thiol-exchange and DNA adsorption onto bare Au willbe characterized with iSEFMI and then compared to DNA SAMs manufactured while applyinga constant potential. Potential-assisted thiol-exchange and DNA adsorption onto clean goldduring the application of a square wave modulated potential (ESWdep ) will be performed and the1076.2. Experimentalresulting surfaces compared. As in previous section, the influence of surface crystallographywill be examined. The influence of the DNA concentration in the deposition solution will bestudied and analysis of the resulting DNA SAMs will aid our mechanistic understanding of thepotential-assisted DNA deposition while applying a square-wave modulated potential.6.2 ExperimentalThe procedure for potential-assisted deposition was similar to the one presented in the pre-vious chapter. The most significant difference is the application of a 50 Hz square wavepotential (ESWdep ) instead of a constant potential (Edep). The duration of MCH and DNA de-positions, rinsing procedures, electrolyte and DNA concentrations used were identical for boththiol-exchange and DNA adsorption on clean Au (5.2). The IB in this chapter is the same as theprevious chapter, containing a 10 mM TRIS pH 7.5 buffer, 100 mM NaCl and 500 mM MgCl2.The deposition apparatus is slightly different as a counter electrode was added to increase thetemporal response of the system (Figure 6.1). This was to ensure that the time constant of thecell is significantly shorter than the length of the potential step in the 50 Hz square wave (10ms). The counter electrode was a Au wire modified with MCH so as to prevent DNA adsorp-tion. This counter electrode was immersed in the DNA solution with a portion of it covered bya teflon tube until the wire was to prevent electrical contact with the working electrode.Before applyingESWdep , an initial characterization, of either the MCH SAM (for thiol-exchange)or the bare Au electrode (for DNA adsorption on clean gold), was performed using EIS at OCP.The EIS measurement involved applying the measured OCP with the addition of a 10 mVpeak to peak sinusoidal perturbation covering a frequency range from 30 kHz to 10 Hz. Afterfitting the EIS measurements to an RC circuit, a capacitance was determined. ESWdep , whichconsists of a 50 Hz square wave with the positive limit at +0.4V/SCE or +0.5 V/SCE and thenegative limit at -0.3 V/SCE (Figure 6.2), was applied for the duration of the deposition time (5minutes). The electrode was disconnected from the potentiostat, removed from the depositionsolution and then either rinsed with Millipore water before storage in IB overnight (for the thiol-exchange case) or immersed for 90 minutes in a 1 mM MCH solution in IB (the final step ofDNA adsorption on clean Au).1086.2. ExperimentalFigure 6.1: Schematic depicting electrodeposition set up for 2 electrode system used in theprevious chapter (a) and the 3 electrode system used to apply ESWdep (b)For comparison, potential-assisted thiol-exchange and potential-assisted DNA adsorptionon clean gold during application of a constant applied potential (Edep) for 5 minutes was alsodone. The procedures for manufacturing these DNA SAMs are identical to what was done inthe previous chapter (Section 5.2).-0.4-0.2 0 0.2 0.4 0.6 0  30  60  90  120E (V/SCE)Time (ms)Figure 6.2: Potential profile applied during DNA deposition at ESWdep (50 Hz square wave, +0.4Vto -0.3V/ SCE) applied during the DNA immersion time (5 minutes).1096.3. Results and Discussion6.3 Results and Discussion6.3.1 Potential-Assisted Thiol exchange with Square-Wave ModulatedPotentialA comparison of the influence of Edep vs ESWdep is presented first for the DNA thiol-exchangeprocess. In the previous chapter, applying a constant Edep = +0.4V/SCE for 60 min duringDNA thiol-exchange resulted in a higher DNA coverage due to an increase in the number ofdefects with which the alkanethiol modified DNA interacts. Creating a similar DNA coverageon gold surfaces with shorter immersion times may be possible using square-wave depositionpotential perturbations.147,192,193 To test this, a comparison of the surfaces prepared during a5 min application of either Edep (+0.4V / SCE) or ESWdep (50 Hz square wave, +0.4V to -0.3V/SCE), was performed on MCH SAM modified single crystal gold bead electrodes during DNAthiol-exchange. The resulting modified surfaces were analyzed using iSEFMI and the fluores-cence images are shown in Figure 6.3a and b with the measured fluorescence intensities fromregions containing low index planes, (111), (100) and (110), and selected high index planes,(311) and (210), in Figure 6.3c. The estimated DNA coverage that corresponds to the averagefluorescence intensity is about 1.2 × 1012 DNA/cm2, which is 10-20% of its maximum packingdensity.15Overall, the fluorescence intensity of the DNA SAMs made using Edep or ESWdep appear tobe very similar. The fluorescence intensities for 5 selected regions show that the underlyingsurface crystallography has a more significant influence on the extent of thiol exchange. Thelargest DNA surface coverage is on the (311) surface. This is a result of this region having aMCH SAM of weakly adsorbed thiols or due to the presence of more defects.19 This contrastswith the (210) surface which has significantly lower DNA coverage as a result of the strength ofthe adsorption of MCH to the (210) surface.19 Interestingly, comparing the constant Edep andESWdep methods result in surfaces with similar coverages for these five regions. The (111) facetshowed a significant difference when using a Student’s t-test but only at an 80% confidenceinterval (see Appendix A for calculation). This suggests that applying ESWdep does not result in asubstantially higher DNA coverage overall, but some discernible differences were noted. Thefluorescence intensity pattern in the regions between the (100) facet and (311) or the (210)1106.3. Results and DiscussionFigure 6.3: Fluorescence images obtained through iSEFMI of DNA SAMs prepared usingpotential-assisted DNA thiol-exchange (for 5 min) at (a) Edep = +0.4V/SCE and (b) ESWdep (50Hz square wave, +0.4V to -0.3V/ SCE). A stereographic triangle is used to label five surfacecrystallographies with the measured averaged fluorescence intensities on select surface crys-tallographies shown in (c). The errorbars shown in the bar graphs correspond to the largerof either the standard deviation of at least 2 replicates or the propagated measurement er-rors. Reprinted with permission from Leung and colleagues.169 Copyright (2019) AmericanChemical Society.regions are more defined for ESWdep which may reveal differences in the nature of the createddefects. The application of a negative potential step near the reduction of the thiol bondedto the gold surface should destabilize or create defects in the SAM, thereby enhancing thiol-exchange. This seems to be specific for particular surface crystallographies as revealed by thefluorescently labeled DNA. Using ESWdep does impact the extent of the surface that experiencesthiol-exchange, in particular the regions around the (111) facet. This suggests that usingESWdep results in accelerated thiol-exchange characteristics which can be explained through theformation of defects, or weakening of the gold thiol bond. It may also be due to an increase inthe local concentration of DNA in the surface region due to pulsing as described previously147which can be investigated by using a lower [DNA] in the deposition buffer.The evolution of the DNA surface coverage on the various regions of the electrode as afunction of [DNA] in the deposition solution may also reveal surface specific thiol-exchangecharacteristics enabling a detailed comparison of Edep (+0.4V) or ESWdep (+0.4V to -0.3V/SCE,50Hz) (Figure 6.4a and b, respectively). The fluorescence intensities from the same crystallo-graphic regions for this range of [DNA] are shown in Figure 6.5. The DNA coverage increasesas expected with increasing [DNA] for both deposition methods. The relative fluorescence1116.3. Results and DiscussionFigure 6.4: Fluorescence images of DNA SAMs prepared via DNA thiol-exchange (for 5 min)at a) Edep = +0.4V/SCE and (b) ESWdep ( +0.4V to -0.3V/SCE, 50 Hz) for varying concentrationsof DNA in the deposition solution. Reprinted with permission from Leung and colleagues.169Copyright (2019) American Chemical Society.intensities from each crystallographic region remain the same but the overall fluorescence in-tensity (i.e. DNA coverage) increases. The DNA coverage is slightly higher for the 0.25 and 0.5μM DNA for ESWdep especially on and around the (111) facet as described previously; though theoverall DNA coverage does not significantly increase when using ESWdep . While control over theDNA coverage can also be realized by choosing the appropriate [DNA] in the deposition solu-tion, the DNA coverage is more influenced by the choice of the applied potential or potentiallimits as will be examined in the next section.6.3.2 Positive Potential Limit and Potential-Assisted DNA Thiol-ExchangeIn the last chapter, it was shown that using a potential more positive than +0.4V/SCE dur-ing a 60 min thiol-exchange results in an higher DNA coverage because of the creation of ahigher density of defects in the MCH SAM. Using a higher positive potential and comparing1126.3. Results and DiscussionFigure 6.5: Average fluorescence intensity for five selected regions representing low in-dex planes (111), (100), (110) and high index planes (210), (311) on DNA SAMs made atEdep=+0.4 V/SCE (filled points) and ESWdep = +0.4 V to -0.3 V/SCE (50Hz) (closed points). Eachdata point correspond to an individual replicate measurement. Reprinted with permission fromLeung and colleagues.169 Copyright (2019) American Chemical Society.1136.3. Results and DiscussionEdep=+0.5V/SCE, 5 min and ESWdep ( +0.5V to -0.3V/ SCE, 50 Hz, 5 min) is shown in Fig-ure 6.6 with a comparison of the fluorescence intensities for the thiol-exchange depositionmethods from the 5 regions of interest given in Figure 6.7. Comparing electrodeposition atEdep=+0.4V/SCE and +0.5V/SCE shows a modest increase in DNA coverage for most of theregions studied with a significant increase observed for the (110) region. The increase in theESWdep positive potential limit also resulted in a DNA coverage increase (compared with ESWdep(0.4 V to -0.3V/SCE)) except for the (311) that has a lower coverage. When comparing theESWdep treated samples, the largest increases were observed for the (111), (110), and (210) re-gions. Even though the variability between samples was large, significant differences are stillobserved including in the regions which were not specifically analyzed. For example, for theESWdep (0.4 V to -0.3V) samples, low DNA coverage was observed for regions around the (100),specifically between the (100)-(110) and (100)-(111) crystallographic zone axes. Increasingthe positive potential limit significantly increased the DNA coverage in these regions, specifi-cally along the (100)-(210)-(110) zone. This is also the case when analyzing the regions thatsurround the (111) facet, which are similar to small (111) terraces separated by many steps.182In agreement with the results from the previous chapter, an increase in the positive electrode-position potentials results in greater MCH SAM instability and facilitated DNA thiol-exchange.When compared to the constant Edep condition, using ESWdep does not dramatically improve theDNA coverage, and in some cases decreases the coverage. This can be explained since thetime spent at negative potentials, where the stability of the MCH SAM is greater, limits the timespent at the positive potential where the MCH SAM is less stable resulting in a lower DNAcoverage due to a lower number of defects created in the MCH SAM.DNA thiol-exchange during the application of Edep or ESWdep manipulates the creation of de-fects in the MCH SAM which is strongly dependent on the underlying surface crystallography.These studies also demonstrate the significant role which the positive potential limit has on thecreation of defects and the extent of thiol-exchange with more positive electrodeposition po-tentials favoring an increase in the DNA coverage. Using either Edep or ESWdep conditions doesnot seem to significantly impact the DNA coverage and therefore the thiol-exchange process.Interestingly, the surface modified at Edep=+0.5V/SCE for 5 min is similar to the surfacecreated after modification at Edep=+0.4V/SCE for 30 min, shown in the previous chapter (Fig-1146.3. Results and DiscussionFigure 6.6: Fluorescence images obtained through iSEFMI of DNA SAMs prepared with DNAthiol-exchange (for 5 min) at a) Edep = +0.5V/SCE and (b) ESWdep ( +0.5V to -0.3V/ SCE, 50 Hz)with a stereographic triangle to label five surface crystallographies. c) measured fluorescenceintensities for these crystallographic surfaces. The errorbars shown in the bar graphs corre-spond to the larger of either the standard deviation of at least 2 replicates or the propagatedmeasurement errors. Reprinted with permission from Leung and colleagues.169 Copyright(2019) American Chemical Society.ure 5.7). In both cases, the DNA diffusion to the electrode surface will be similar as theelectrophoretic effect is negligible in high ionic strength buffer, which suggests that the morepositive potential destabilizes or weakens the gold-thiol interaction, significantly increasing theextent of thiol exchange. The high Cl– concentration present may play a role in destabiliz-ing the gold-thiol interaction which will explored in the next chapter. Literature has shown aninfluence of thiols on the anodic oxidation of gold in Cl– containing electrolyte with Au disso-lution starting at +0.6V/Ag|AgCl.194 DNA coverage is not enhanced with applied square-wavepotentials as compared to constant potential, but rather it is influenced by the applied positivepotentials in Cl– containing electrolyte.6.3.3 Square-Wave Modulated Potential-Assisted DNA Adsorption on a CleanAu BeadTypically preparation of DNA biosensors involves adsorption of thiol-modified DNA onto a cleanAu surface at OCP followed by displacement of non-specifically adsorbed DNA by MCH.91The use of a square wave deposition potential was employed147,193 during DNA adsorptionon clean gold and the resulting surfaces were characterized with EIS in the presence of1156.3. Results and DiscussionFigure 6.7: Comparison of fluorescence intensity for DNA SAM thiol exchange for five selectedregions prepared at OCP, Edep (+0.4V or +0.5V) and ESWdep (+0.4 V to -0.3 V/SCE (50Hz) or+0.5 V to -0.3 V/SCE (50Hz)). The errorbars shown in the bar graphs correspond to thelarger of either the standard deviation of at least 2 replicates or the propagated measurementerrors. Reprinted with permission from Leung and colleagues.169 Copyright (2019) AmericanChemical Society.Fe(CN) 3 – /4 –6 . This characterization method indirectly measures the extent of DNA adsorp-tion through the defects in the DNA SAM. The authors reported a higher DNA coverage for theSAMs prepared using 15 min of potential pulsing in phosphate buffers with 0.45M K2SO4 ascompared to OCP and compared to SAMs prepared while applying a constant potential.147 Inthe previous chapter, DNA adsorption onto clean gold with and without applying a constant po-tential resulted in DNA SAMs with high coverage. Applying Edep = +0.4 V/SCE or Edep = +0.5V/SCE only slightly increased the DNA coverage. Based on the results of Jambrec and col-leagues, we expected to observe an increase in DNA coverage when applying a square-wavemodulated potential.Here, a comparison of the modified gold bead surfaces made at OCP, or using either Edep= +0.4 V/SCE or ESWdep = +0.4 V to -0.3 V/SCE (50Hz) during DNA adsorption onto clean goldfor 5 minutes was performed in a TRIS buffer containing 1.1 M Cl– . Fluorescence images ofDNA SAM modified surfaces (followed by immersion in MCH for 90 min) are shown in Figure6.8a-c with fluorescence intensities from select crystallographic features compared in Figure6.8d. The difference between surfaces prepared using Edep or ESWdep methods does not appearto be significant though the ESWdep method results in slightly higher fluorescence intensity (orDNA coverage). The (111) and (311) regions show a statistically significant increase in fluo-rescence intensity when analyzing the results using a Student’s t-test with a confidence levelof 80% (see Appendix A). DNA adsorption for surfaces prepared using ESWdep appear to prefer1166.3. Results and DiscussionFigure 6.8: Fluorescence images of DNA SAMs prepared on clean Au (for 5 min) at a) OCP;b)Edep =+0.4 V/SCE; c) ESWdep = +0.4 V to -0.3 V/SCE (50Hz) followed by immersion in 1mMMCH in IB for 90 min.; d) Measured fluorescence intensities for five selected DNA SAMs onthese surface crystallographies. A stereographic triangle is shown labeling the five surfacecrystallographies. d) measured fluorescence intensities for these crystallographic features.The errorbars shown in the bar graphs correspond to the larger of either the standard deviationof at least 2 replicates or the propagated measurement errors. Reprinted with permission fromLeung and colleagues.169 Copyright (2019) American Chemical Society.the regions around the (100) facet and the stepped regions surrounding the (111) facet. Theseregions are like miscut (111) or high index planes, similar to the substrate defects (e.g., grainboundaries) that would be present in vapor deposited gold film on glass or polycrystalline goldelectrodes. Therefore, square-wave potential assisted deposition could significantly increasethe average DNA coverage on these conventional substrates. Since the coverage of DNA isclose to the maximum expected on this surface, differences in the DNA coverage may be dif-ficult to observe. Further measurements using a lower concentration of DNA in the depositionbuffer were also performed and may reveal differences in DNA adsorption on clean gold atconstant applied potential and square-wave modulated potential.Fluorescence images of gold beads modified with DNA SAMs prepared using potential-assisted deposition onto a clean Au bead surface with increasing concentrations of thiol mod-ified DNA in the deposition solution are shown in Figure 6.9a and b for Edep and ESWdep re-spectively. Both Edep and ESWdep methods result in the same increase in DNA coverage withincreasing [DNA] in the deposition buffer. No significant differences in fluorescence intensitywere observed between the two deposition methods for the range of concentrations studiedfor the five surface crystallographic regions selected (Figure 6.10).1176.3. Results and DiscussionFigure 6.9: Fluorescence images of DNA SAMs made with DNA adsorption onto a clean Aubead electrode surface (for 5 min) using a) Edep =+0.4 V/SCE and (b) ESWdep = +0.4 V to -0.3V/SCE (50Hz) with increasing concentrations of DNA in the deposition buffer. Reprinted withpermission from Leung and colleagues.169 Copyright (2019) American Chemical Society.1186.3. Results and DiscussionFigure 6.10: Measured fluorescence intensities on the low index surfaces (111), (100) and(110) and high index planes (311) and (210) for DNA SAMs at varying DNA concentrationswith 5 minute DNA adsorption at Edep=+0.4V/SCE (closed points, solid line) and ESWdep = +0.4V to -0.3 V/SCE (50Hz) (open points, dotted line). Reprinted with permission from Leung andcolleagues.169 Copyright (2019) American Chemical Society.1196.3. Results and Discussion6.3.4 Positive Potential Limit and DNA Deposition onto Clean AuAlso studied was the deposition of DNA onto a clean gold bead electrode using a more positivepotential. Edep =+0.5 V/SCE and ESWdep = +0.5 V to -0.3 V/SCE (50Hz) were applied for 5 min toclean gold electrodes in the DNA deposition buffer. After MCH treatment, the resulting surfaceswere compared based on the fluorescence images (shown in Figure 6.11 d and e). Depositionat Edep = +0.5V results in a generally uniform and high DNA coverage SAM (comparison isgiven in Figure 6.12) with lower coverage on the (311), as seen for Edep=+0.4V depositions.While an increase in fluorescence can be observed depending on the positive potentials used,ESWdep did not result in a substantial increase in the DNA coverage. This may be a consequenceof the saturation of fluorescence intensity with increasing DNA coverage (see section 4.2.4);this unfortunately decreases the ability to measure the influence of the deposition potential forregions with a high DNA coverage. Other groups used this type of square-wave modulatedpotential assisted DNA deposition on planar polycrystalline gold substrates and demonstrateda higher coverage than compared to OCP deposition based on EIS measurements.147,193This conclusion is not evident using our fluorescence imaging approach, illustrating that eachcharacterization technique seems to be sensitive to different aspects of the modified surface.While EIS is sensitive to defects and pinholes in the adsorbed layer, the fluorescence imagingis powerful for examining the distribution of coverage, especially in the lower surface coveragerange. Nevertheless, our results can be used to explore the mechanism proposed for thisenhanced DNA deposition when using ESWdep which has been suggested to be due to ion stirringeffect (increasing the transport of DNA to the surface).1936.3.5 Charging of the Interface during Potential-Assisted DNA DepositionThe application of ESWdep during DNA deposition appears to affect the assembly of DNA on thegold bead surface in a similar manner to applying Edep whether performed on clean gold orthiol modified gold surfaces as measured using fluorescence. ESWdep modulates the surfacecharge density which should manipulates the movement of ions near the electrode surface.This was proposed to be important in explaining the increased DNA coverage in addition tousing potential limits that are on either side of the pzc.147 They proposed that in the high1206.3. Results and DiscussionFigure 6.11: Fluorescence images of DNA SAMs prepared on clean Au (for 5 min) at a)OCP; b)Edep =+0.4 V/SCE; c) ESWdep = +0.4 V to -0.3 V/SCE (50Hz); d) Edep =+0.5 V/SCE; e)ESWdep = +0.5 V to -0.3 V/SCE (50Hz), followed by immersion in 1mM MCH in IB for 90 min.A stereographic triangle is shown labeling the five surface crystallographies. d) measuredfluorescence intensities for these crystallographic features. Reprinted with permission fromLeung and colleagues.169 Copyright (2019) American Chemical Society.Figure 6.12: Measured fluorescence intensities for DNA SAM manufactured via adsorption onclean Au for five selected surface crystallographies prepared at OCP, Edep (+0.4V or +0.5V)and ESWdep (+0.4 V to -0.3 V/SCE (50Hz) or +0.5 V to -0.3 V/SCE (50Hz)). The errorbars shownin the bar graphs correspond to the larger of either the standard deviation of at least 2 repli-cates or the propagated measurement errors. Reprinted with permission from Leung andcolleagues.169 Copyright (2019) American Chemical Society.1216.3. Results and Discussionionic strength buffer, the potential only affects the region near the electrode surface, and if thepotential limits used for the square wave were centered around the pzc, there would be anexchange of cations and anions causing stirring effect which transports DNA to the surfaceincreasing coverage. Since the pzc for clean gold is a function of crystallography, significantdifferences in DNA coverage enhancement should be observed over the surface of the goldbead electrodes used in the present study for potential-assisted DNA SAM formation on cleangold.DNA deposition onto clean gold surface did not show an obvious correlation with the pzc,which suggests a different mechanism is at play. Furthermore, this can be supported by exam-ining the amount of charge that is passed during the ESWdep method for the clean gold surfaceand a MCH SAM covered gold surface, given that they have significantly different capacitances(~40 and 2 µF/cm2 respectively).The current during ESWdep was measured (at 5kHz) at the beginning and end of the 5 mindeposition (Figure 6.13) and the net charge passed through the electrode during DNA deposi-tion via thiol-exchange and adsorption on a clean gold surface are shown in Figure 6.14. Thedeposition solution was not free of O2 and may result in a slow negative drift in the chargedepending on the surface coverage. The total magnitude of charge that moves through theinterface for the MCH modified surface is significantly smaller (about 5%) when compared tothe modification of the clean gold surface (Figure 6.15), as expected since the capacitance islower. If charge modulation and ion stirring is causing an enhancement of DNA deposition,then DNA SAM formation on clean gold during application of Edep vs ESWdep would be signifi-cantly compared to DNA thiol-exchange during application of Edep vs ESWdep . This is not whatwas observed using our fluorescence imaging method in a high [Cl– ] buffer.1226.3. Results and Discussion-0.20-0.15-0.10- 0  30  60  90  120  150  180Current (mA)Time (ms)DNA Thiol-Exchanget=0 s t=300 s(a)-10.00- 0  30  60  90  120  150  180Current (mA)Time (ms)DNA Adsorption on Clean Au t=0 s t=300 s(b)Figure 6.13: Current measured over 0.2 seconds at the start and end of the potential-assisteda) DNA Thiol-exchange and b) DNA adsorption on Bare Au while applying an ESWdep = +0.4 V to-0.3 V/SCE-4.0- 0  30  60  90  120  150  180σM (µC/cm2 )Time (ms)DNA Thiol-Exchanget=0 s t=300 s(a)-20.0- 0  30  60  90  120  150  180σM (µC/cm2 )Time (ms)DNA Adsorption on Clean Au t=0 s t=300 s(b)Figure 6.14: Charge measured over 0.2 seconds at the start and end of potential-assisteda) DNA Thiol-exchange and b) DNA adsorption on clean Au while applying an ESWdep = +0.4V to -0.3 V/SCE Reprinted with permission from Leung and colleagues.169 Copyright (2019)American Chemical Society.1236.4. Conclusions0102030 0  30  60  90  120  150  180|σ M| (µC/cm2 )Time (ms)DNA Thiol-Exchange(a)0100200300400500 0  30  60  90  120  150  180|σ M| (µC/cm2 )Time (ms)DNA Adsorption on Clean Au (b)Figure 6.15: Magnitude of charge passed to the gold bead electrode interface during the first0.2s of the 5 min electrodeposition using ESWdep for DNA SAMs created using a) thiol-exchangemethod or b) DNA adsorption onto clean gold followed by MCH. Calculated from the datashown in Figure 6.14. Reprinted with permission from Leung and colleagues.169 Copyright(2019) American Chemical Society.6.4 ConclusionsDifferent potential-assisted methods for rapid preparation of DNA SAMs (5 min) on singlecrystal gold bead electrodes were examined with in-situ spectroelectrochemical fluorescencemicroscopy for a consistent and reliable comparison. The use of a constant Edep or square-wave ESWdep for both thiol-exchange and direct assembly of DNA SAMs on clean gold surfaceswere studied. The application of a positive potential significantly influenced the DNA SAM cov-erage as compared to not applying a potential (OCP) for the thiol-exchange process. The DNAcoverage clearly depended on the surface crystallography and all regions except for the (311)were strongly influenced by the application of potential. No significant difference was observedwhen comparing constant and square-wave potential deposition methods for negative poten-tials limited to -0.3V/SCE and positive potentials limited to +0.4V/SCE. The largest changeswere observed when using a more positive potential (for either the constant potential or thepositive potential limit of the square wave), which resulted in higher DNA coverage on the sur-face, and obvious dependence on the underlying substrate crystallography. The mechanismproposed to explain an improvement in DNA coverage through ion stirring was not supportedin these measurements which may be due, in part, to the use of a supporting electrolyte that1246.4. Conclusionscontains a large [Cl– ]. A defect mediated mechanism can uniformly explain the formation ofthe high coverage DNA SAMs and the influence of the applied potentials, providing insightinto the merit of electrodeposition of DNA SAMs for a facilitated, better controlled biosensorsurface preparation.125Chapter 7The Influence of Chloride Anions inthe Electrolyte During DNA Deposition7.1 IntroductionThe previous chapters described DNA SAMs made with potential-assisted thiol-exchange andpotential-assisted DNA adsorption onto clean gold while applying a constant potential (Edep)or a square-wave modulated potential (ESWdep ). It was established that the applied potentialincreases the number of defects in the SAM, resulting in enhanced DNA thiol-exchange andDNA SAM formation on clean gold. The DNA depositions presented earlier were done in animmobilization buffer (IB) containing 1.1 M Cl– which may play a significant role in increasingthe number of defects and therefore leading to increased DNA coverage in the DNA SAM.Through STM measurements, the adsorption of alkanethiols and disulphides onto goldwhile applying a potential was observed to be enhanced in the presence of chloride.195 To un-derstand this phenomena, the interaction of chloride with the metal surface must be discussed.Chloride is known to form a chemical bond with the gold, silver and mercury surfaces. Thisbond involves a electronic charge from the anion redistributed towards the metal surface.196As previously explained in the chapter 2, specifically adsorbed ions, like chloride, reside in theinner Helmholtz layer and as a result of their lack of a solvation shell, are able to form high cov-erage adlayers on the metal surface. In addition, the presence of specifically adsorbed anionsresults in the metal surface having a more negative pzc. Chloride adsorption has been char-acterized on Au(111) using electrochemistry189 and X-Ray diffraction34, and also on Au(100)through in-situ scanning tunneling microscopy (STM) measurements.197 Shi and Lipkowskistudied the electrochemical behaviour of chloride adsorption on gold at potentials above -0.21267.1. IntroductionV/SCE. They measured a pzc for Au(111) at 0.2 V/SCE and noted that gold oxidation shiftedto more positive potentials in the presence of 1 mM Cl– .189 The authors also measured amaximum chloride coverage of 9 × 1014 ions/cm2 (at 0.9 V/SCE in 5 mM KCl).189 From -0.2V, the chloride adlayer was found to become more densely packed with increasingly positivepotential. A phase transition occurs near ~0.7 V/SCE, where the chloride adlayer changesfrom being disordered to tightly-packed hexagonal structures.34,197In addition to adsorption, chloride has been observed to induce Au dissolution at positivepotentials.191,198–201 This dissolution is given in equation 7.1 and was observed to occur at 1.1V/SCE in a 1 mM Cl– solution.191,198Au + 4Cl− −−→ AuCl −4 + 3e− (7.1)On Au(111), when increasing Cl– concentration, the dissolution rate increases and the on-set of this dissolution was found to shift to less positive potentials.191 The dissolution, observedwith STM, was found to occur anisotropically, as a result of the organization of the chloride ad-layer on Au(111). Furthermore, as modeled with Monte Carlo simulations and supported byin-situ STM measurements, Au surface atoms with adsorbed chloride were found to be moremobile compared to other Au atoms at the surface.190,199 These studies on the adsorption ofchloride on gold and the subsequent dissolution of gold were all done in presence of millimolarquantities of Cl– . In the current work, 1.1 M Cl– was present in the DNA immobilization buffer,which is three orders of magnitude greater than previous studies. The influence of chlorideon the gold surface will most definitely be observed as seen in the previous studies. Hence,changes are expected for DNA SAM made when no chloride is present in the electrolyte duringthe potential-assisted DNA thiol-exchange and DNA SAM formation on clean gold.In this chapter, DNA SAMs made via potential-assisted thiol-exchange and potential-assistedDNA SAM formation on gold will be done with reduced amounts or complete removal of chlo-ride. First, differential capacitance on a MCH layer measurements in a chloride-free bufferwill be done in order to explore differences in the number of defects created by the presenceof chloride. This will be followed by examining DNA SAMs made via potential-assisted thiol-exchange in electrolyte containing reduced amounts of chloride. As adsorbed chloride plays a1277.2. Experimentalrole in the surface charge and the dissolution of gold, potential-assisted DNA SAM formationon clean gold will be studied in chloride-free immobilization buffers. Both DNA SAM formationduring application of a constant potential (Edep) and a square-wave modulated potential (ESWdep )without chloride will be shown. The resulting DNA SAMs will be characterized based on theirDNA coverages and their local environment. These DNA SAMs will also be compared to DNASAMs made in the same manner but with chloride, as was done in previous chapters.7.2 ExperimentalThe procedure for making potential-assisted DNA SAMs in the absence of chloride were similarto that shown in previous chapters. Both DNA thiol-exchange and DNA SAM formation onclean Au were done as detailed in section 5.2. Application of Edep and ESWdep was done usingthe 2-electrode and 3-electrode systems, described in sections 5.2 and 6.2. In this chapter,changes were made to the immobilization buffer (IB) in which the DNA deposition occurs.The salt bridge used to connect the reference electrode to the DNA solution was filled withan appropriate salt solution corresponding to the IB composition. For MCH immersion, whichoccurs as the second step for DNA SAM formation on clean gold, the MCH was diluted in abuffer that was either the same or similar to the IB used during the DNA deposition. For DNAthiol-exchange, MCH solutions were always diluted in MeOH. Four different IB solutions withpH 7.5 are used and their composition are detailed below:The IB containing Cl– detailed in previous chapters was used again. This buffer contained10 mM TRIS 100 mM NaCl and 500 mM MgCl2. For DNA SAM formation on clean gold, MCHsolutions were diluted in the same Cl– containing IB. As before, the salt bridge used duringthe potential-assisted DNA deposition was filled with 0.1 M NaCl.An IB containing TRIS with no Cl– was used and had 10 mM TRIS, 100 mM KNO3 (≥99.0%BioXtra, Sigma Aldrich) and 500 mM Mg(NO3)2 (>98.0% Sigma Aldrich) in lieu of NaCl andMgCl2 respectively. 1 mM MCH solutions used during DNA SAM formation on clean gold werediluted in the same NO –3 containing IB and the salt bridge used during the DNA depositionwas filled with 0.1 M KNO3. Unless specified in the text, the pH of this TRIS buffer was adjustedusing HNO3 instead of TRIS HCl salt.1287.3. Results and DiscussionA phosphate IB used was composed of 10 mM NaH2PO4/Na2HPO4 (Monobasic, Fisher99% enzyme grade; DibasicFisher Anhydrous ACS grade) and 450 mM Na2SO4 (ACS Anhy-drous). 1 mM MCH solutions were diluted into buffer containing 10 mM NaH2PO4/Na2HPO4and20 mM Na2SO4. The salt bridge used during the DNA deposition was filled with 0.450 MNa2SO4.An IB containing phosphate buffer with Cl– was composed of 10 mM NaH2PO4/Na2HPO4and 1 M NaCl. The salt bridge used during the DNA deposition was filled with 1 M NaCl. 1mM MCH solutions were diluted in a buffer containing 10 mM NaH2PO4/Na2HPO4and 50 mMNaCl.7.3 Results and Discussion7.3.1 Influence of Chloride on a Mercaptohexanol SAMIn the previous chapters, the IB used during potential-assisted thiol-exchange contained ahigh [Cl– ] (1.1 M). Chloride is known to chemisorb to gold surfaces34,189,197 and dissolve Au atpositive potentials.191,198–201 Although not many studies have been done on the role of chlorideon alkanethiol SAMs, Zamborini and colleagues have studied alkanethiol SAMs in solutions10 mM Br– in 0.1 M HClO4 using FTIR-external reflection spectroscopy and STM.202 Whenapplying potentials >0.75 V/SCE, Br– was found to corrode the gold through pinholes in theSAM, eventually resulting in defects in the SAM.Here, the effect of Cl– on an alkanethiol SAM is characterized by comparing differentialcapacitance measurements of a MCH coated gold bead electrode in an electrolyte containingTRIS, NaCl and MgCl2 (Figure 7.1, magenta) and in a chloride free electrolyte with TRIS,KNO3 and Mg(NO3)2 (Figure 7.1, blue). The potential is scanned from 0 V/SCE to negativepotentials in Figure 7.1a and from 0 V/SCE to positive potentials in Figure 7.1b. The onset ofreductive desorption of the MCH layer occurs at the same potential (<-0.6V/SCE) in both theCl– and non Cl– electrolyte. This suggests that Cl– does not affect the reductive desorptionprocess. Interestingly, a rise in the capacitance between -0.2 V to -0.6 V is observed only whenCl– is present. In Chapter 5, we attributed this to the wetting of the electrode. With no previous1297.3. Results and DiscussionFigure 7.1: Capacitance of a MCH SAM on the gold bead electrodes measured starting at0V and (a) scanning to negative potentials or (b) scanning to positive potentials revealing theinfluence of Cl– in the stability of the MCH SAM. Adapted from Leung and colleagues.168Copyright (2018) with permission from Elsevier.studies of Cl– on gold surfaces at potentials below -0.2 V/SCE, it is difficult to speculate onwhy wetting would only happen with Cl– present. A comparison of the capacitance measuredfor the MCH coated gold bead for high and low [Cl– ] at positive potentials is shown in Figure7.1b. In the presence of Cl– the onset of oxidative desorption is shifted to a less positivepotential. Additionally, the capacitance increases at potentials close to 0.4 V/SCE, indicatingthat defects that form in the SAM can be attributed to the presence of Cl– . Therefore, the DNASAMs, made through potential-assisted thiol-exchange while applying a positive Edep or ESWdep ,likely have enhanced DNA coverage due to defects created by the presence of Cl– .7.3.2 Influence of Chloride on a Potential-Assisted DNA Thiol-ExchangeThe influence of Cl– on DNA thiol-exchange while applying Edep= +0.40 V/SCE was inves-tigated using an IB with lower [Cl– ]. The NaCl was replaced with KNO3 and MgCl2 wassubstituted with Mg(NO3)2 to keep the ionic strength and [ Mg2+] constant such that the DNAinteradsorbate interaction does not change. The TRIS component of the buffer contributes8 mM of chloride (due to the use of TRIS HCl as the conjugate acid) as does the DNA stock1307.3. Results and Discussion(a) (b)Figure 7.2: a) Fluorescence images of thiol-exchanged MCH/DNA SAMs prepared by elec-trodeposition at Edep= +0.40 V for 1 hour in 1 μM DNA diluted in immobilization buffers witha large [Cl– ] or a significantly lower [Cl– ]. b) Fluorescence intensities on select surface crys-tallographic regions are shown. Of the five surface crystallographies measured, removing asignificant amount of chloride Cl– results in a decrease in the DNA coverage on (110). Adaptedfrom Leung and colleagues.168 Copyright (2018) with permission from Elsevier.solution resulting in 28 mM Cl– in the electrolyte.Figure 7.2a compares the DNA SAMs made via DNA thiol-exchange at Edep = +0.40 V/SCEfor 60 min in 1 μM DNA solutions diluted in the original IB with [Cl– ] = 1.1 M and in an IB com-posed of 10 mM TRIS, 0.1 M KNO3, 0.5 M Mg(NO3)2 (with [Cl– ] = 28 mM). Comparing high andlow [Cl– ] reveals a significant decrease in fluorescence intensity seen in the region surround-ing the (100) region and the (110). The (311), (100), (111), and (210) regions are not stronglyaffected, while the (110) is lower in intensity (Figure 7.2 b). The high [Cl– ] does influence thethiol-exchange for certain high index regions around the (100) and thereby facilitates a uni-form deposition of DNA. These results also suggest that Cl– acts to create defects (similar towhat was proposed for Br– 202) where thiol-exchange can occur more readily on all crystallo-graphic surfaces (except for the (210) at Edep = +0.4 V/SCE which has a low DNA population).These results show that a high [Cl– ] is beneficial for creating uniform surface modification viapotential-assisted thiol-exchange on most crystallographic features. This approach bodes wellfor more complex surfaces such as vapor deposited gold films on glass which have a varietyof low and high index grains and grain boundaries.The corrosion of gold by specifically adsorbed anions such as Cl– and Br– is known to bepassivated by an alkanethiol SAM.194,202 Therefore, when performing potential-assisted DNA1317.3. Results and DiscussionSAM formation on clean Au, where no initial MCH SAM is present to impede Au dissolution,the presence of Cl– may have a greater effect on the DNA SAM formation process.7.3.3 Potential-Assisted DNA Adsorption on Clean Au in ChlorideTo understand the influence of Cl– on potential-assisted DNA SAM formation on clean gold,the DNA SAMs made with application of Edep= +0.4V or ESWdep = +0.4V/-0.3V (50 Hz), as waspresented in previous chapters, must be reexamined. In particular, a focus on the local envi-ronment of the DNA SAM is necessary, as it affects the hybridisation capabilities of the DNAstrands adsorbed to the surface.In Chapter 5 (5.3.6), characterizing the DNA local environment was done by measuring thefluorescence modulation in relation to the fluorescence intensity. The fluorescence modulation,which is the change in the fluorescence intensity as a result of actuating the DNA electrostat-ically (via applying the potential profile shown in section 4.2.2 Figure 4.6), is related to thelocal environment surrounding the DNA. With increasing DNA interadsorbate interactions, theangular motion of the DNA becomes restricted, leading to a decrease in fluorescence modu-lation. As established previously, for a DNA SAM on a given crystallographic region where theDNA are homogeneously spaced, the fluorescence modulation decreases as the maximumfluorescence intensity, and its corresponding DNA coverage, increases. The experimentallydetermined maximum fluorescence modulation was empirically fitted to the data in Chapter 5and was drawn as a dotted line in Figure 5.15. The maximum fluorescence modulation wasalso calculated in section 5.3.6 for two models: one where the angular motion of DNA be-comes restricted while the DNA tilts in the same direction as the neighbouring DNA (ModelA) and another where the angular motion of DNA becomes restricted when the DNA tilt to-wards one another (Model B) and is shown in Figure 5.14. At a given fluorescence intensityand DNA coverage, fluorescence modulations which are lower than that of a homogeneouslyspaced DNA SAM are non-ideally distributed. The mobility of the DNA can only be reducedby interactions with other DNA at the surface. With the number of DNA adsorbed remainingthe same, the lower fluorescence modulation can only be caused by DNA adsorbing in a dif-ferent configuration. DNA SAMs made via potential-assisted thiol-exchange at all potentials1327.3. Results and Discussionwere previously proven to be homogeneous, indicated by the proximity of the measured fluo-rescence modulation to the experimentally determined maximum fluorescence modulation.Fluorescence images of DNA SAM formation on clean gold while applyingEdep= +0.4V/SCEand ESWdep = +0.4V/-0.3V (50 Hz) are shown in Figure 7.3a. When depositing DNA from a so-lution containing 0.5 μM DNA for 5 minutes, applying Edep= +0.4V/SCE results in DNA SAMsthat have higher fluorescence intensity, therefore have a larger DNA coverage compared tothe ESWdep treatment. As explained in the previous chapter, this is due to differences in defectscaused by the applied potentials. Applying positive potentials creates a greater number ofdefects in certain regions compared to negative potentials. While applying ESWdep , only half ofthe 5 minute deposition time is spent at +0.4V/SCE, resulting in a lower number of defects andlower DNA coverage.During the iSEFMI characterization, where the DNA is actuated by the modulated poten-tial profile, the fluorescence intensity of the DNA SAM was monitored. For DNA SAMs onthe Au(111) and Au(100) made via DNA adsorption of clean gold with application of Edep=+0.4V/SCE or ESWdep = +0.4V/-0.3V (50 Hz), the fluorescence intensity changes while the DNAare actuated (Figure 7.3a). In agreement with the relation between the fluorescence mod-ulation and the maximum fluorescence intensity, larger DNA coverages result in a smallerchanges in the fluorescence intensity . A comparison of the fluorescence modulation wasplotted against the maximum fluorescence intensity for DNA SAMs on five regions on the elec-trode surface (each associated with a different surface crystallography: (111), (100), (110),(210), (311)) is shown in Figure 7.3b. DNA SAMs presented here were made from solutionswith [DNA] ranging from 0.1 μM to 1 μM. The measurements here are distinguished based onthe potential applied during the DNA SAM formation (Edep or ESWdep ). The black dashed linerepresents the maximum fluorescence modulation that was observed in all experimental mea-surements regardless of the method used to prepare the DNA SAM surfaces. In addition, thecalculated fluorescence modulation from the two DNA movement limiting conditions (Model Aand B) are shown as green dashed lines.The surfaces prepared in TRIS buffer with NaCl and MgCl2 follow the expected trend asthe DNA coverage increases. DNA SAMs made while applying Edep showed better agreementwith the experimentally determined maximum fluorescence modulation, suggesting that these1337.3. Results and DiscussionDNA SAMs contain DNA that is homogeneous spaced. DNA SAMs made while applying aESWdep perturbation had slightly lower fluorescence modulation than the Edep case for the samesurface coverage of DNA. This suggests that some of the DNA in the SAM are adsorbed witha different configuration such that they are not able to move as freely, resulting in a slightlylower fluorescence modulation. Exploring the effect of the electrolyte composition during theDNA SAM formation may shed some light on their discrepancies.7.3.4 Potential-Assisted DNA Adsorption on Clean Au without ChloridePreviously, the presence of Cl– in the immobilization buffer during potential-assisted DNA thiol-exchange was shown to be important for preparing more uniform DNA SAMs. The potential-assisted DNA SAM formation on clean gold was done using an immobilization buffer withthe same [Mg2+] as before but Cl– was replaced with NO –3 so as to keep the ionic strengthsimilar. The fluorescence images of the resulting DNA SAMs are shown in Figure 7.4a. Theabsence of Cl– did not significantly change the DNA surface coverage for DNA SAMs madewith application of Edep, but a slight decrease in DNA coverage was realized for DNA SAMsmade when applying ESWdep .The SAMs on Au(111) and Au(100) prepared while applying Edep show a high degree ofDNA reorientation or motion as indicated by the fluorescence modulation, whereas the SAMsprepared while applying ESWdep results in a low coverage DNA SAM with an order of magnitudelower fluorescence modulation (Figure 7.4a). The fluorescence intensity and the lower fluo-rescence modulations observed were consistent for replicate measurements, indicating a lackof hysteresis. The results suggest that the DNA is restricted in its mobility by its neighbors, somuch that the fluorescence modulation is lower than that the calculated maximum fluorescencemodulation for model B (Figure 7.4b). Deposition with an applied Edep creates DNA SAMsthat have a more homogeneous distribution. The fluorescence modulation behaviour on DNASAMs made at ESWdep can be explained by the presence of high surface density DNA clustersdispersed across the surface (Figure 7.5). These clusters consist of DNA that are adsorbedclose together despite the availability of other surface sites. Scanning electron microscopy im-ages (shown in appendix C) supports the formation of such clusters. Further characterization1347.3. Results and DiscussionFigure 7.3: Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdepfor 5 minutes in a 0.5 μM DNA solution in TRIS IB containing Cl– . During iSEFMI measure-ments, modulated potential profile was applied with a fluorescence image taken at each step.The fluorescence intensity on (111) and (100) surfaces during this measurement is shown withgreen data points indicating the intensity during the positive potential excursions and black dat-apoints indicating negative potential excursions. (c) The measured fluorescence modulationplotted over the maximum fluorescence intensity for DNA SAMs on all other surfaces madeat different [DNA] (0.1 μM to 1 μM) while applying Edep or ESWdep for 5 minutes. The maximumfluorescence modulation for a homogeneously spaced DNA SAM is shown from experimentaldata, and from calculations based on Model A and Model B. The top x axis indicates the DNAcoverage, correlated from measurements made in Results and DiscussionFigure 7.4: Fluorescence images of DNA SAMs made while applying (a) Edep or (b)ESWdep for5 minutes in a 0.5 μM DNA solution in TRIS IB containing NO –3 . The fluorescence intensitywas measured on (111) and (100) during the iSEFMI measurements. The changes in the fluo-rescence intensity as a result of DNA reorientation are smaller for the DNA SAMs made whileapplying ESWdep . (c) The measured fluorescence modulation plotted over the maximum fluores-cence intensity for DNA SAMs on all other surfaces made at different [DNA] (0.25 μM to 0.5μM ) while applying Edep or ESWdep for 5 minutes. The fluorescence modulation on DNA SAMsformed while applying ESWdep deviated from the experimentally determined fluorescence modu-lation for an ideal homogeneously spaced DNA SAM, as well as the calculated fluorescencemodulation for the highly restrictive model B.1367.3. Results and DiscussionFigure 7.5: Schematic depicting the formation of high surface density clusters for two DNASAMs with the same DNA coverage. The top schematic depicts the DNA SAM made whileapplying Edep results in homogeneous non-clustered DNA SAMs and the bottom depicts theDNA SAM made while applying ESWdep containing high density clusters.with atomic force microscopy, or Förster resonance energy transfer would enable confirma-tion of these DNA clusters. Cluster formation is plausible based on the assumption that themaximum fluorescence intensity is measured when the DNA is standing up. It is possible thatthe SAM made using ESWdep is organized such that the DNA is unable to become fully uprightwhen performing the iSEFMI measurements. This would also result in a lower fluorescencemodulation. With the current methods of measurement, it is difficult to distinguish whether theDNA SAM is organized in this way or if the DNA SAM contains clusters. In further discussions,the results presented here are proposed to be a result of DNA forming clusters.One could surmise that DNA cluster formation occurs as follows: after one DNA moleculeadsorbs to the surface, another DNA molecule adsorbs immediately next to the initial adsorbedDNA, instead of another site on the surface. Subsequent DNA prefer to adsorb next to ad-sorbed DNA resulting in multiple islands of adsorbed DNA instead of homogeneously spacedDNA. Interestingly, clusters were not present in DNA SAMs made while applying Edep in Cl–or in NO –3 . Meanwhile, cluster formation is prominent when using ESWdep conditions in the ab-sence of Cl– . It appears that the cluster formation is strongly curtailed in Cl– , suggesting animportant role for Cl– in this formation of DNA SAMs while applying ESWdep . Cl– is known toadsorb to the gold surface,196 and has also been reported to increase the mobility of the gold1377.3. Results and Discussionatoms on the surface.190,199 This may result in the DNA adsorbates electrodeposited via ESWdepinitially forming clusters but later becoming distributed by the mobilized gold as a result of Cl– .Although it was observed that the application of ESWdep during potential-assisted DNA ad-sorption on clean gold with no Cl– present results in DNA cluster formation, it is important tonote the presence of Mg2+. As this divalent cation reduces the negative repulsion betweenphosphate groups in the DNA backbone,82 the high Mg2+ in the TRIS containing IB could aidcluster formation. Assessing the same DNA deposition in a different electrolyte compositioncontaining no Mg2+ is needed to further understand the factors which affect cluster formation.7.3.5 Potential-Assisted DNA Adsorption on Clean Au in Phosphate BufferOther than TRIS containing buffers, the preparation of DNA SAMs has been done with DNAsolution diluted in phosphate buffers.15,89 Potential-assisted DNA SAM formation was doneon clean gold but in a phosphate buffer (with added Na2SO4 to ensure similar ionic strength).Additionally, the electrolyte used is identical to that used by Jambrec and colleagues in theircomparison of DNA SAMs made at Edep or ESWdep .147 Fluorescence images of the resultingDNA SAM made while applying Edep and ESWdep are shown in Figure 7.6. The DNA surfacedensity is substantially smaller than the DNA SAMs made in the TRIS IB. This is expectedgiven the absence of divalent cations which acts to decrease the electrostatic repulsion duringimmobilization. To minimize the electrostatic repulsion between the adsorbed DNA molecules,the DNA should be sparsely dispersed on the surface., Hence during iSEFMI measurements,the DNA should have a high degree of reorientation due to the modulated potential applied.DNA SAMs prepared while applying Edep have a higher DNA coverage than seen for DNASAMs made while applyingESWdep . This is in agreement with the results observed for DNA SAMsmade with TRIS containing IB, indicating that the presence of phosphate does not change therelative effect of ESWdep and Edep on the DNA coverage (Figure 7.3a and 7.4a). Though theconditions were similar to that used by Jambrec and colleagues, the DNA coverage measuredvia iSEFMI characterization of DNA SAMs on single crystal bead electrodes contradicts theirfindings.147 EIS measurements were attempted on the DNA SAMs similar to those performedby Jambrec and colleagues. These measurements yielded irreproducible signals suggesting1387.3. Results and Discussionthat this method was not ideal for characterizing DNA SAMs on single crystal bead electrodes.Even with these challenges, no differences were observed in the signals between DNA SAMsmade at Edep with ESWdep .Similar to the DNA SAMs made in the TRIS IB containing no Cl– , the local environmentaround the tethered DNA is dramatically different when comparing the Edep and ESWdep elec-trodeposition procedures. Though the overall DNA coverage is lower in DNA SAMs madewhile applying ESWdep , a significant decrease in DNA fluorescence modulation was observed.Comparing the fluorescence modulation measured for DNA SAMs on all the surface crystal-lographic features measured (Figure 7.4b), the fluorescence modulation measured on DNASAMs formed during application of ESWdep are below that of model b. This indicates that amongthe DNA SAMs formed, with certain exceptions, the reorientation capability of the DNA is re-duced indicating DNA clusters. DNA SAMs made on different surface crystallographies whileapplying Edep have higher fluorescence modulations, indicating that the DNA is able to re-orient with the modulating potential. Except at lower coverages, the measured fluorescencemodulation is close to the fluorescence modulation for a more uniformly distributed DNA SAM.Despite the lack of Mg2+ in the immobilization buffer, the DNA appears to form clusters as aresult of ESWdep . This suggests that the cluster formation is not enhanced by the divalent cationsin the electrolyte. With only a monovalent cation (Na+) present, the applied ESWdep leads to DNAadsorption such that the neighbouring DNA strands overcome the negative repulsion and areable to adsorb next to each other even at low DNA coverages.The influence of Cl– in cluster formation is investigated again via DNA SAM formation whileapplying ESWdep using a phosphate IB with Cl– . The fluorescence images of the DNA SAMscreated are shown in Figure 7.7a. The substitution of Na2SO4 with NaCl resulted in DNASAMs that were of reasonable coverage. Though the Na+ concentration in the phosphate IB isthe same as before, the DNA coverage appears to be higher when the DNA electrodepositionoccurs in the presence of Cl– . With no Mg2+ present, the enhancement in the DNA coverageis likely a result of the increased number defects due to adsorption of Cl– on the gold surface.The movement of the DNA, based on the fluorescence modulation, did not appear completelyrestricted in DNA SAMs made with application of ESWdep . It is clear in Figure 7.7b, that thereare substantially less DNA SAMs with measured fluorescence modulations that are below1397.3. Results and DiscussionFigure 7.6: a) Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdepfor 5 minutes in a 0.5 μM DNA solution in a phosphate IB containing Na2SO4. The fluores-cence intensity was measured on (111) and (100) during the iSEFMI measurements. Thefluorescence modulations are smaller for the DNA SAMs made while applying ESWdep suggest-ing cluster formation. (c) The measured fluorescence modulation plotted over the maximumfluorescence intensity for DNA SAMs on all other surfaces made at different [DNA] (0.1 μM to1 μM) while applying Edep or ESWdep for 5 minutes. The fluorescence modulation on DNA SAMsformed while applying ESWdep deviated from the experimentally determined fluorescence modu-lation for an ideal homogeneously spaced DNA SAM, as well as the calculated fluorescencemodulation for the highly restrictive model B.1407.3. Results and Discussion10% (compared to the fluorescence modulation of DNA SAMs deposited with applied ESWdepwithout Cl– ). However, the fluorescence modulation still deviates from the ideal fluorescencemodulation for a homogeneously spaced DNA.Based on these results, the presence of Cl− inthe phosphate IB reduces the formation of clusters during the SAM formation. The role of Cl–is clearly important in preventing DNA cluster formation, while the role of Mg2+ is less importantfor cluster formation. This is surprising since the presence of Mg2+ significantly reduces theelectrostatic DNA-DNA adsorbate interactions but from these results does not have a biggerrole to play in the extent of DNA cluster formation.7.3.6 Analysis of DNA SAM Local Environment for Different SurfaceCrystallographiesThe size of the DNA clusters is likely on the nm scale, as they could not be optically charac-terized directly using a 40x water immersion objective. Scanning probe microscopy or super-resolved fluorescence microscopy measurements would be required to directly investigate fac-tors that affect the cluster formation. For now, iSEFMI is used to indirectly study the extent ofcluster formation (fclusters) as a result of the DNA SAM formation conditions. This can be doneby analyzing the deviation of fluorescence modulation from that of a non-clustered DNA SAM(Equation 7.2). Since the experimental fluorescence modulation is lower than the ideal, thedeviation is positive.Fl.mod.ideal − Fl.mod.expFl.mod.ideal× 100% = Perc.Dev. = fclusters (7.2)The percent deviation of the fluorescence modulation from the ideal fluorescence modu-lation of a homogeneously spaced DNA SAM is characterized by the shaded areas in Figure7.8. The deviation represents the percentage of clustered DNA that has restricted reorientationability (or a fluorescence modulation of zero) in a DNA SAM on a given region. Rearrangementof equation 7.2 shows that this percent deviation relates to the extent of clusters formation.Fl.mod.exp = Fl.mod.ideal(100− fclusters) (7.3)1417.3. Results and DiscussionFigure 7.7: Fluorescence images of DNA SAMs made while applying (a) Edep or (b) ESWdep for 5minutes in a 0.5 μM DNA solution in a phosphate IB containing NaCl. The fluorescence inten-sity was measured on (111) and (100) during the iSEFMI measurements. The reorientation ofthe DNA , measured during iSEFMI, is not restricted for the DNA SAMs made while applyingESWdep . (c) The measured fluorescence modulation plotted over the maximum fluorescence in-tensity for DNA SAMs on all other surfaces made at different [DNA] (0.25 μM to 0.5 μM) whileapplying Edep or ESWdep for 5 minutes. The fluorescence modulation measured on DNA SAMsmade while applying ESWdep only slightly deviated from the experimentally determined fluores-cence modulation for an ideal homogeneously spaced DNA SAM. This suggests that clusterformation still occurs but is dampened by the presence of Cl–1427.3. Results and DiscussionThe dashed curve indicates the fluorescence modulation for homogeneously spaced DNASAMs. This deviation can be used to evaluate the DNA SAMs made as a result of the deposi-tion conditions during potential-assisted DNA SAM formation (Figure 7.9). DNA SAMs madewhile applying Edep with an IB containing Mg2+ had the least deviation from ideal DNA mobility(<50% clustered DNA SAMs). The DNA SAMs made in a Cl– IB deviate by 20% from idealDNA SAM behaviour and DNA SAMs made with no Cl– present deviate by 30% from idealDNA SAM behaviour. Significant deviation from the ideal is evident in DNA SAMs made whileapplying ESWdep in the absence of Cl– . The average percent deviation from the ideal of theseDNA SAMs were high (~75% clustered surfaces) as a result of a large number of high densityclusters. The presence of Cl– when performing potential-assisted deposition at ESWdep signifi-cantly reduces cluster formation in the DNA SAM. DNA SAMs made while applying ESWdep in theTRIS Cl– IB on average had ~30% clustered DNA on the surface. For DNA SAMs made whileapplying ESWdep in the phosphate Cl– IB on average had ~45% clustered DNA on the surfacethough this varied across different surfaces. This behaviour can be explained by the favourableformation of DNA clusters with application of ESWdep . Furthermore, the presence of Cl– in theimmobilization buffer results in the dispersion but not full elimination of these clusters.The local environment of the DNA SAM can be further analyzed as a function of surfacecrystallography made possible by the use of the single crystal gold bead electrodes. In the pre-vious figures, DNA SAMs on the surface crystallographies (111), (100), (110), (311) and (210)were represented together in the fluorescence modulation vs. fluorescence intensity plots. InFigure 7.10, the DNA SAMs, prepared in TRIS and phosphate buffers without Cl– , on each ofthe individual five surface regions are compared. As shown in the results above, DNA clustersform with application of ESWdep in the absence of Cl– . The data in Figure 7.10 reveal that allof the surfaces studied, the low index planes, (111) and (100), show the largest prevalence ofthese clustered DNA organizations. Some clustering is noted for the other surface regions, butthe largest deviations are for the low index planes (111) and (100) when on DNA SAMs madewhile applying ESWdep . This differences in cluster formation on each surface crystallography islikely due to the mechanism of SAM formation. The formation of alkanethiol SAMs has beenproposed to proceed via nucleation and growth.51–53 For DNA SAM formation, the number ofnuclei could be larger on the more open faces such as the (110), (311) and (210) due to the1437.4. ConclusionFigure 7.8: Fluorescence modulation vs. fluorescence intensity plot with shaded areas wherethe measured fluorescence modulation deviates from the experimentally determined maximum(dashed line). Measured fluorescence modulation that falls on the dashed line indicates anideal homogeneously distributed DNA SAM. Measured fluorescence modulation which is onthe respective shaded region indicates that the given percentage of this DNA SAM is clustered.Select data points from Figure 7.4 are plotted here for demonstration.larger step density on these surfaces and higher atomic roughness. As nuclei tend to formonly on defect sites as found in atomically rougher surfaces, a low density of nucleation sitesis expected on atomically smoother surfaces such as (111) and (100) surfaces. As a result, onthese low index planes, the DNA adsorbs on the surface around the few nucleation site leadingto the growth of DNA clusters.Comparing the phosphate buffer (without Mg2+) and TRIS buffer (with Mg2+) shows thatthe dication does not dramatically influence the prevalence of clusters, suggesting the ionicstrength of the deposition buffer solutions (>1.35 M for Phosphate IB with 0.45M Na2SO4) wassufficient for minimizing electrostatic repulsions between neighboring DNA adsorbates.7.4 ConclusionThe composition of the IB used during potential-assisted DNA thiol-exchange and potential-assisted DNA SAM formation on clean Au, was shown to have an influence on the DNA SAMformed. In potential-assisted thiol-exchange, the presence of Cl– , which specifically adsorbs1447.4. Conclusion 0 1 2 3 4 529.3+/-13.3	N=10FrequencyDC TRIS KNO3- 0 2 4 6 76.4+/-14.3	N=20FrequencySW TRIS KNO3- 0 5 10 15 17.0+/-21.7	N=40FrequencyDC TRIS Cl- 0 5 10 0  20  40  60  80  10033.2+/-30.3	N=40FrequencyPercent Deviation(%)SW TRIS Cl- 0 2 4 6 8 32.8+/-13.3	N=25FrequencyDC PB SO42- 0 5 10 1575.3+/-15.0	N=35FrequencySW PB SO42- 0 2 4 6 20.6+/-19.0	N=20FrequencyDC PB Cl- 0 1 2 3 4 5 0  20  40  60  80  10045.5+/-23.8	N=20FrequencyPercent Deviation(%)SW PB Cl-Figure 7.9: The percent deviation from the ideal behaviour of a homogeneously spaced DNASAM for DNA SAMs made while applying Edep and ESWdep in their respective IB. The percentdeviation represents the portion of the DNA SAM which is composed of high density DNAclusters.1457.4. Conclusiona) Phosphate Buffer without Cl-100101102102 103 1041012 1013{111}Fl. Mod. (%)Fl. Int. (counts)PB no Cl-DCSW102 103 1041012 1013{100}Fl. Int. (counts)102 103 1041012 1013{110}Fl. Int. (counts)102 103 1041012 1013{311}Fl. Int. (counts)102 103 1041012 1013{210}Fl. Int. (counts)b) Tris buffer without Cl-100101102102 103 1041012 1013{111}Fl. Mod. (%)Fl. Int. (counts)TRIS w/o Cl-DCSW102 103 1041012 1013{100}Fl. Int. (counts)102 103 1041012 1013{110}Fl. Int. (counts)102 103 1041012 1013{311}Fl. Int. (counts)102 103 1041012 1013{210}Fl. Int. (counts)Figure 7.10: Fluorescence modulation vs. fluorescence intensity plots for DNA SAMs on theirrespective crystallographic surfaces: (111), (100), (110), (210) and (311). SAMs were de-posited while applying either Edep or ESWdep in a) Phosphate Buffer without Cl– or (b) Tris bufferwithout Cl–onto gold, creates defects at applied potentials in the initial MCH SAM, enabling more DNAthiol-exchange. As a result, the DNA SAM formed via thiol-exchange while applying Edep=+0.4V/SCE in Cl– creates a uniform DNA SAM across the entire single crystal bead electrode.Potential-assisted DNA SAM formation on clean gold was performed with application ofa constant potential (Edep=+0.4 V/SCE) or modulated potential (ESWdep =+0.4 V to -0.3 V/SCE(50 Hz square wave)) in two different buffers: with and without Cl– . These DNA SAMs werethen analyzed using iSEFMI to characterize the DNA coverage and the local environmentaround the DNA. A modulated potential was used to induce reorientation of the fluorophorelabeled DNA adsorbed on the surface. Changes in the fluorescence intensity were used toindicate steric restrictions of the adsorbed DNA due to a neighbouring DNA. The use of asingle crystal bead electrode also facilitated measuring the influence of surface crystallographyon the resulting DNA SAMs.DNA SAM formation on clean gold from a TRIS buffer containing 1 M Cl– resulted in DNASAMs consisting of DNA that were uniformly distributed with little preference to the underlyingcrystallography. In this buffer, the use of a Edep or ESWdep resulted in very similar DNA modified1467.4. Conclusionelectrode surfaces. The adsorbed DNA displayed a decrease in the extent of DNA reorientationdue to the modulated potential as the DNA coverage increased. This was consistent with thebehaviour of a DNA SAM composed of homogeneously distributed DNA. The nature of theDNA SAM changed when performing the DNA deposition while applying ESWdep in the absenceof Cl– . The surfaces contained a significant amount of DNA on the surface that could notreorient with the modulated potential, even when the DNA coverages were low. The localenvironment around the DNA in the SAM appeared to be highly constrained, suggesting theformation of DNA clusters in the SAM. Performing the DNA deposition while applying ESWdep ina phosphate buffer without Cl– also resulted in DNA SAMs with clusters. With DNA depositionat ESWdep performed in phosphate buffer with Cl– , a significant reduction of these DNA clusterswas observed.The influence of dication species, Mg2+, did not have any effect on the presence or absenceof clusters. The formation of the DNA clusters is significantly reduced in the presence of Cl– .This was explained to occur due to the mobility that specifically adsorbed Cl– provides to theAu surface atoms. DNA bound to these gold atoms would also be mobile, reducing the clusterformation. The clusters, when present, were most prevalent on the low index plane surfaces(111 and 100) and significantly less so on the higher index planes like (210) or (311). Thissuggests that the formation of the DNA SAM via electrodeposition occurs via nucleation andgrowth, where the number of nucleation sites depends on the number of steps or defects onthe surface. The few nucleation sites on the atomically smoother 111 and 100 surfaces resultsin preferential growth around these sites resulting in the formation of DNA clusters in the SAM.The presence of adsorbed Cl– either creates more nucleation sites, or provides enhancedmobility to the adsorbed alkanethiol modified DNA, preventing cluster formation.The serendipitous discovery of these DNA clusters formed as a result of DNA depositionwhile applying ESWdep highlights the need for further characterization of DNA SAMs i.e. notlimiting characterization to DNA coverage measurements alone. Further exploration may beneeded in order to properly elucidate a mechanism for cluster formation and to further iden-tify factors which affect characteristics such as cluster size and spacing. Sensitive techniquessuch as atomic force microscopy, scanning tunneling microscopy and super-resolved fluores-cence microscopy may allow direct measurement of these clusters. For now, using iSEFMI1477.4. Conclusioncharacterization, the conditions for creating DNA clusters or preventing clusters in a DNA SAMare outlined. The possibility of these DNA clusters having increased hybridisation capabilitieshas been suggested.159 Furthermore, the ability to control cluster formation can be used tofurther tailor DNA SAMs to be used for biosensing or in creating novel biomaterials.203148Chapter 8Conclusion and Future Outlook8.1 Summary of ResultsWhen DNA SAMs are used as DNA biosensors, characteristics such as DNA coverage andthe local environment of the DNA are crucial to their performance. The DNA coverage on thesurface as well as the local environment surrounding each DNA are parameters that affect howthe DNA can bind to complementary strands or biomolecules of interest. These DNA SAMsrequire an optimized manufacturing procedure where its characteristics are controlled and theDNA SAM is made in a fast and facile manner. Ma and Lennox previously demonstrated thatapplying a positive potential during alkanethiol self assembly resulted in alkanethiol SAMscreated in a short time and with fewer defects.14 This was later shown for DNA SAMs withPeterson and colleagues where applying positive potentials resulted in high coverage DNASAMs formed in a shorter time period.15In this thesis, the formation of fluorescently labeled DNA self-assembled monolayers (SAMs)on gold surfaces while applying a potential (Edep) has been described and the resulting DNASAMs were characterized through fluorescence microscopy. Two of the procedures used toform DNA SAMs were modified to include an applied potential during the DNA depositionstep. The first procedure involved DNA thiol-exchange with a MCH layer89 and the secondprocedure consisted of DNA SAM formation on clean gold followed by the displacement ofnon-specifically adsorbed DNA with MCH.103 The influence of the applied potentials on theresulting DNA SAMs were investigated in addition to DNA adsorbate concentration and DNAdeposition time.91,96 Additionally, as the gold surface morphology is known to affect the config-uration of adsorbates on the surface,11 the influence of surface crystallography was explored.This was done by forming DNA SAMs on single crystal gold bead electrodes whose surfaces1498.1. Summary of Resultswere comprised of a predetermined arrangement of surface crystallographies. The use of asingle crystal bead electrode enabled observation of DNA SAM formation over different sur-faces in a self-consistent fashion as was done previously for alkanethiol SAMs and DNA SAMsmade without applied potential.19 Furthermore, in doing this, the development of a procedurewhich enhances the DNA coverage for all surface crystallographies in a short time span couldbe done. Studies like this would be difficult to conduct on commercial gold substrates. Thesesurfaces are polycrystalline and have an uncontrolled variety and composition of surface crys-tallographies.The fluorescently labeled DNA SAMs formed on single crystal bead electrodes were char-acterized using in-situ electrochemical fluorescence microscopy imaging (iSEFMI). With thefluorescence images of the DNA SAMs, the DNA were directly probed and heterogeneousfeatures such as DNA aggregation and preferential adsorption could be observed. Theseheterogeneous features are not visible with EIS, SPR and other averaged measurement tech-niques typically used to evaluate DNA SAMs.12Potential-assisted thiol-exchange method and potential-assisted DNA SAM formation onclean gold was performed while applying a constant potential (Edep) or a modulating potential(ESWdep ) and characterized using iSEFMI. Comparisons of DNA SAM formation in different elec-trolyte compositions were also studied in order to understand the role of the adsorbing anionswith applied potentials. A summary of the findings of each experimental chapter within thisthesis is provided below:Potential-assisted DNA thiol-exchange with application of a constant positive Edep for man-ufacturing DNA SAMs was investigated. DNA SAMs made via DNA thiol-exchange with aconstant positive Edep applied were compared to SAMs made without an applied potentialbased on fluorescence images of the DNA SAMs. iSEFMI characterization revealed that DNAthiol-exchange at a constant positive Edep yielded DNA SAMs with higher DNA coverage. Aspreviously demonstrated with EIS measurements of alkanethiol SAMs,170–173 defects in theMCH SAM increased with positive potentials. These potential-induced defects in the initialMCH SAM facilitated the thiol-exchange of the MCH SAM with DNA. The extent of defectcreation in the MCH SAM was found to depend on the underlying surface crystallography; de-fects were found to increase on low index planes with increasingly positive applied potentials.1508.1. Summary of ResultsOn high index planes, the number of defects depended on the surface. The MCH SAMs onAu(311) had numerous defects regardless of potential and MCH SAMs on Au(210) only con-tained defects above a threshold positive potential. Differences in the DNA SAM formed viathiol-exchange were observed for those made at positive Edep and those made at negativeEdep. These differences were likely related to the presence of chloride, which induces moredefects in the MCH SAM at positive potentials. Nevertheless, the local environment of theDNA in the SAM, probed via iSEFMI measurements, was found to be similar regardless of thepotential applied during the DNA deposition. It was concluded that the DNA were homoge-neously spaced on the surface, with higher DNA coverages resulting in more tightly-packedDNA.Potential-assisted thiol-exchange was also compared to potential-assisted DNA SAM for-mation on clean gold while applying a constant Edep. It was found that after a short depositiontime, the DNA could adsorb at high coverages. No major surface crystallography dependencewas observed except for the (311) surface, which had a lower DNA coverage. Applying Edepduring DNA SAM formation on clean gold appeared to result in a slight increase in the overallDNA coverage. However, the changes in the DNA coverage were small in comparison to therange of the coverages observed for DNA SAMs made via potential-assisted thiol-exchange.DNA SAMs made via DNA thiol-exchange and DNA SAM formation on clean gold duringapplication of a modulated square wave potential (ESWdep ) were compared to DNA SAMs formedwhile applying a constant Edep. The comparison was done in order to confirm the hypothesisfound in the literature where applying ESWdep resulted in higher coverage DNA SAMs with a shortdeposition time compared to applying a constant Edep.147 It was important to ensure that thefindings in previous studies applied to all surface crystallographies, and not just the surfaceswhich dominate on the polycrystalline substrates previously used.It was found that during DNA thiol-exchange with application of ESWdep , compared to Edep,did not significantly increase the overall DNA coverage nor did it increase the DNA coverageon five surface crystallographies measured. However, enhanced DNA coverage was observedin terraces surrounding the 111 region thought to be found in abundance on polycrystallinesurfaces used typically. The small differences in the DNA coverages supports the theory thatapplying either Edep or ESWdep during DNA thiol-exchange results in defects in the MCH SAM1518.1. Summary of Resultscharacteristic to the applied potential. Increasing the positive limit of the ESWdep or increasingthe Edep applied during DNA thiol-exchange resulted in significantly larger increases in theDNA coverage. This supports the results found earlier where increasingly positive potentialscreated more defects in the MCH SAM, resulting in a greater extent of thiol-exchange anda higher DNA coverage. In other words, adjusting the positive potential limit affects the DNAcoverage more than simply including a negative potential excursion as is done when applying amodulated potential (ESWdep ). ApplyingESWdep instead ofEdep during DNA SAM formation on cleangold also did not result in an enhanced DNA coverage. Similar DNA coverages were observedon the five surface crystallographies of interest when comparing DNA SAMs made with ESWdepto DNA SAMs made with Edep. Applying ESWdep resulted in small increases in the DNA coveragein certain regions similar to that found when applying ESWdep during DNA thiol-exchange. It wasmore difficult to observe any enhancement of the DNA coverage for potential-assisted DNAdeposition on clean gold due to the high DNA coverage.Previous literature suggested that the ion movement due to ESWdep was pertinent to the en-hanced DNA coverage.147,193 This would suggest that an enhanced DNA coverage is expectedfor DNA SAM formation on clean gold while applying ESWdep instead of Edep. Despite the largeramount of ion movement during DNA SAM formation on clean gold compared to during DNAthiol-exchange, more DNA coverage enhancement was seen on DNA SAMs made with DNAthiol-exchange when applying ESWdep instead of Edep. Enhancements in the DNA coverage dueto ESWdep observed could only be a result of defects being created in the DNA SAM as a resultof the applied potential and not dictated by the ion-stirring caused by the modulated potential.Finally, the influence of an adsorbing anion in the electrolyte used during the potential-assisted DNA thiol-exchange and DNA deposition on clean gold was explored. Supporting theprevious literature, adsorbing anions such as chloride create more defects in the MCH layer atpositive potentials.196,202 As a result, DNA thiol-exchange, while applying a positive constantEdep in the presence of chloride, resulted in higher DNA coverages. In comparison to DNASAMs prepared with less chloride, these DNA SAMs had more consistent coverages acrossdifferent surface crystallographies. The high chloride concentration present during potential-assisted thiol-exchange would enable uniform DNA distribution on polycrystalline gold elec-trodes except on certain grain boundaries.1528.1. Summary of ResultsPotential-assisted DNA SAM formation on clean gold while applying Edep or ESWdep in theabsence of chloride was examined. The local environment of the DNA was probed duringiSEFMI characterization through which the ability of the DNA on the surface to reorient wasmeasured. It was found that the DNA SAMs made while applying ESWdep had low coverageand contained DNA whose ability to reorient was restricted by neighbouring DNA. This be-haviour could only be explained with the DNA adsorbing into clusters of high packing densityon the surface. The formation of these DNA clusters was minimized by the presence of chlo-ride, likely due to the increased mobility of the Au atoms as a result of chloride adsorption.Potential-assisted DNA SAM formation on clean gold was also performed without magnesiumin a phosphate buffer. With fewer divalent cations present to shield the repulsion betweenthe negatively charged DNA, the resulting DNA SAMs had lower DNA coverages compared tothose made in the original TRIS buffer containing Mg2+. Without Cl– , DNA clusters still formed,unaffected by the removal of the divalent cation. A greater extent of DNA clusters were foundon atomically smooth low index planes compared to more atomically rough high index planes.Atomically smooth surfaces contain less nucleation for DNA to adsorb, resulting in DNA thatwas non-homogeneously spaced on these low index surfaces.In examining the role of the electrolyte composition, it was clear that adsorbing anionssuch as Cl– play a major role in the electrodeposited DNA SAMs by creating defects in theMCH layer during thiol-exchange, or preventing cluster formation during DNA SAM formationon bare gold. The use of Mg2+ as a divalent cation, which shields the negative charge on theDNA,82 affects the overall DNA coverage but does not significantly influence cluster formation.The work in this thesis has demonstrated that a number of parameters influence the DNASAMs made as a result of potential-assisted deposition. The DNA coverage on certain surfacecrystallographies can be controlled depending on the choice of potential (positive or negativeEdep or ESWdep with a chosen positive potential limit) applied during DNA thiol-exchange, dura-tion of DNA deposition or the electrolyte composition. The capability of increasing the DNASAM coverage without having to increase the deposition time to nearly 24 hours significantlyfacilitates the manufacturing process of such DNA SAMs for biosensing purposes. As thecrystallographic composition of polycrystalline surfaces can vary for different electrodes, it wasimportant to find a procedure or an applied potential that enhances the DNA coverage in a1538.2. Future Work and Application to DNA Biosensorsshort time and on all surface crystallographic features. This work highlights the need to ex-amine DNA SAMs on multiple surface crystallographies in addition to those that dominate onpolycrystalline surfaces (e.g. Au(111) and Au(100)).With the eventual use of DNA SAMs as DNA biosensors, the application of DNA SAMsmanufactured using potential-assisted DNA deposition as real sensing devices should be ex-plored. Possible extensions of the current work are shown in the next section, with short termprojects that could further the use of potential-assisted DNA deposition to create DNA biosen-sors.8.2 Future Work and Application to DNA BiosensorsAfter investigation of potential-assisted formation of DNA SAMs, further work utilising thesemethods to manufacture DNA biosensors is required. Prototypes for DNA biosensors havebeen developed which detect the hybridisation of probe DNA assembled on a gold surface witha target complimentary DNA strand from solution.85,111,112,114,115 Additionally, DNA biosensorshave been used to detect biomolecules (e.g. Thrombin) using aptamers assembled onto thesurface.113,133 Electrodeposited DNA SAMs must be tested as a sensor in order to validatethat the applied potential during deposition is beneficial to the DNA sensing performance. Ifthe control of DNA coverage is reproducible using an applied potential, the biosensor can betailored to have an optimal coverage for sensing purposes. This section will briefly describeand summarize future experiments which can be immediately extended from the work in thisthesis.8.2.1 Testing the Hybridisation Efficiency of DNA SAMs made with ControlledCoverageDetecting a target complimentary DNA from solution could be tested using the characteriza-tion methods presented earlier in this thesis. With the electrochemical DNA coverage mea-surements in the presence of RuHex, an observed increase in the measured RuHex adsorbedto the surface signals the detection of the target DNA.37,115,120 The feasibility for hybridisa-tion to occur can be measured through the hybridisation efficiency which can be quantified1548.2. Future Work and Application to DNA Biosensorsvia the percentage increase of RuHex (i.e. DNA) at the surface during hybridisation. As seenpreviously, the hybridisation efficiency is highest at an optimal coverage.15,143 With the elec-trodeposited DNA SAMs created in this thesis, it would be possible to consistently create DNASAMs with this optimal DNA coverage, which would then correspondingly have cosistent hy-bridisation efficiencies. Sequential measurements with iSEFMI, would confirm the amountof fluorescently labeled probe DNA remaining at the surface and additionally reconfirm DNAhybridisation using the method developed by Rant and colleagues.1168.2.2 Potential-Assisted DNA Deposition containing Secondary structuresPotential-assisted deposition of DNA with secondary structures such as hairpin DNA and G-Quadruplex DNA could be investigated. The sensing capabilities of such DNA SAMs contain-ing secondary structures is affected by spacing between the DNA at the surface. For one,the DNA secondary structures require space on the surface to unfold and refold.134 Further-more, having a target analyte that is a protein which can range in size from 1 nm to 100 nm inradius,204 adds more spatial restrictions for detection.Potential-assisted DNA deposition must be tested to create DNA SAMs containing thesesecondary structures. With the DNA coverage control enabled via electrodeposition, a DNAcoverage can be found where the sensing of target biomolecules is best. Testing the de-tection of these target biomolecules with the DNA SAMs can be done with iSEFMI. Usingthe constructs shown by Du and colleagues and Xiao and colleagues,111,113 the fluorescencesignal can be used to indicate the conformation of the DNA SAMs upon binding to the targetmolecule. The fluorescence signal increase is expected when a hairpin probe hybridises with atarget strand,111 and the fluorescence signal is expected to decrease when the G-Quadruplexis formed upon binding to a specific biomolecule.8.2.3 Potential-Assisted DNA Deposition on Polycrystalline Surfaces withImage CharacterizationWhen the sensing capabilities of DNA SAMs assembled on single crystal bead electrode havebeen evaluated, the DNA sensor must be also manufactured and tested using more conven-1558.3. Concluding Remarkstional gold substrates used in commercial biosensors. The information regarding DNA SAMson the different surface crystallographies (from the single crystal bead electrodes) could betranslated to more conventional polycrystalline gold surfaces such as the sputtered gold sub-strates. Most of these polycrystalline surfaces will be made up of low index planes, grainboundaries could have similar morphology to high index planes. DNA SAMs on polycrystallinegold electrodes has previously been imaged using iSEFMI,103 however, this was done withlack of control and understanding of the role of the surface crystallography.Electrodeposited DNA SAMs made on polycrystalline substrates can be characterized us-ing iSEFMI with a few modifications to the spectroelectrochemical cell used. Using a chosenset of conditions, a DNA SAM with optimal DNA coverage and homogeneity can be obtainedon these sputtered gold on glass surfaces. It might be possible, depending on the size of thecrystal domains on the surface and the resolution of the microscope, to discern between theDNA SAMs made on these crystalline domains using iSEFMI.With application of the knowledge on DNA coverage dependence on surface crystallogra-phy, utilising potential-assisted deposition can then be used for immobilizing DNA SAMs ontosputtered gold on glass substrates reproducibly. Using these DNA SAMs as sensors wouldlead to the production of DNA biosensors with sufficient stability and detection capabilities.Subsequently, these upgraded DNA biosensors can be then commercialized and used to ad-vance point-of-care diagnostic systems.8.3 Concluding RemarksDNA biosensors are far from becoming commercialized, as reproducible signals are requiredfor a sensor to be reliable. It is the opinion of the author of this thesis that utilising potential-assisted DNA SAM formation can address the shortcomings of current DNA biosensors. Fur-thermore, the work in this thesis has shown that potential-assisted methods can be a versatilefor tailoring DNA SAMs to a particular biosensor strategy.With the information obtained from iSEFMI analysis, the applied potential was shown tocontrol of the DNA coverage over the different surface crystallographies. The surface crystal-lography affects the heterogeneity of the DNA SAMs built on polycrystalline substrates, which1568.3. Concluding Remarksin turn affects the reproducibility of the DNA SAM. The ability to control the coverage across thedifferent surface features would provide a way of increasing the reliability of the DNA sensor.Reducing the heterogeneity of a DNA SAM built on polycrystalline surfaces can be done bycreating a DNA SAM such that surfaces where the DNA adsorbed have identical local environ-ments. Applying negative potentials during potential-assisted thiol-exchange has been shownto achieve this, resulting in DNA on primarily low-index surfaces (e.g. Au(111)). Alternatively,reducing the heterogeneity can be done by creating DNA SAMs with high coverage on almostall surfaces is possible with DNA adsorption on bare gold. Applying potential during DNAthiol-exchange and DNA adsorption has been shown to affect the DNA coverage on high in-dex regions such as Au(311) and Au(210). As the grain boundaries of polycrystalline surfacesare believed to be high index surfaces,26 it is possible to affect the coverage on these grainboundaries with a given applied potential. Adjustments in the bulk DNA concentration duringpotential-assisted DNA thiol-exchange or DNA SAM formation on clean gold affect the over-all DNA coverage, while keeping the relative DNA coverage over the surface crystallographicfeatures the same.Affecting the DNA coverage on various surface crystallographic features would also enablemodification of nanostructured gold. Geometric surfaces of these nanostructures have uniquesurface crystallography. For example, the ends of gold nanorods have Au(111) surface crys-tallography while the sides are composed of Au(100) and Au(110). By applying an appropriatepotential to these nanostructures during DNA deposition, it may be possible to deposit DNAon specific surfaces to create unique constructs.182With control of the DNA coverage, the DNA sensors can be made such that it is low enoughthat probe DNA have ample spacing for hybridisation to occur or to undergo conformationalchanges.15,144 Additionally, the coverage can also be made high such that mismatches arebetter distinguished or binding of DNA with bulky analytes are hindered.129,205 Furthermore,having low coverage but tightly packed DNA clusters, as is achieved by DNA adsorption onbare gold while applying a square potential waveform without chloride present, may be usefulfor certain biosensor architecture. For example, bivalent analytes which require close packedDNA but at low coverages could be better detected with clustered DNA compared to homoge-neously spaced DNA.2061578.3. Concluding RemarksAs demonstrated, an appropriate DNA SAM manufacturing method can be chosen, de-pending on the eventual use of the DNA SAM with control possible on the overall DNA cover-age, DNA coverage on certain surface crystallographies, and DNA cluster formation. Engineer-ing a setup such that both electrodeposition of the DNA SAM, and electrochemical detectionof a target DNA could occur in the same cell. A flow-through cell could be used to transportdeposition solutions, followed by sample containing target DNA over the metal surface withwashing steps in between. This would be a method of facilitating the manufacturing of theseDNA SAMs and also remove the need to create DNA SAMs with a long shelf-life. Additionally,implementing potential-assisted DNA deposition on an industrial scale could be easily engi-neered to create DNA sensors at a faster rate. With the results of this thesis and the futureresearch to further DNA electrodeposition, it is believed that DNA biosensors can soon be im-plemented commercially which will provide feasible diagnostics and accessible personalizedhealthcare.158References[1] Kelley, S. O. Advancing Ultrasensitive Molecular and Cellular Analysis Methods to Speedand Simplify the Diagnosis of Disease. Acc. Chem. Res. 2017, 50, 503–507.[2] Labib, M.; Sargent, E. H.; Kelley, S. O. Electrochemical Methods for the Analysis ofClinically Relevant Biomolecules. Chem. Rev. 2016, 116, 9001–9090.[3] Huang, Y.; Xu, J.; Liu, J.; Wang, X.; Chen, B. Disease-Related Detection with Electro-chemical Biosensors: A Review. Sensors 2017, 17, 2375–30.[4] Chao, J.; Zhu, D.; Zhang, Y.; Wang, L.; Fan, C. DNA nanotechnology-enabled biosen-sors. Biosens. Bioelectron. 2016, 76, 68–79.[5] Lakhin, A. V.; Tarantul, V. Z.; Gening, L. V. Aptamers: problems, solutions and prospects.Acta naturae 2013, 5, 34–43.[6] Brody, E. N.; Gold, L. Aptamers as therapeutic and diagnostic agents. Reviews in Molec-ular Biotechnology 2000, 74, 5–13.[7] Vallée-Bélisle, A.; Plaxco, K. W. Structure-switching biosensors: inspired by Nature.Curr. Opin. Struct. Biol. 2010, 20, 518–526.[8] Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-AssembledMonolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005,105, 1103–1170.[9] Tang, Y.; Ge, B.; Sen, D.; Yu, H.-Z. Functional DNA switches: rational design and elec-trochemical signaling. Chem. Soc. Rev. 2014, 43, 518–529.[10] Palecek, E.; Bartosik, M. Electrochemistry of Nucleic Acids. Chem. Rev. 2012, 112,3427–3481.159References[11] Ulman, A. Formation and Structure of Self-Assembled Monolayers. Chem. Rev. 1996,96, 1533–1554.[12] Bizzotto, D.; Burgess, I. J.; Doneux, T.; Sagara, T.; Yu, H.-Z. Beyond Simple Cartoons:Challenges in Characterizing Electrochemical Biosensor Interfaces. ACS Sensors 2018,3, 5–12.[13] Gebala, M.; Schuhmann, W. Controlled Orientation of DNA in a Binary SAM as a Keyfor the Successful Determination of DNA Hybridization by Means of ElectrochemicalImpedance Spectroscopy. ChemPhysChem 2010, 11, 2887–2895.[14] Ma, F.; Lennox, R. B. Potential-Assisted Deposition of Alkanethiols on Au: ControlledPreparation of Single- and Mixed-Component SAMs. Langmuir 2000, 16, 6188–6190.[15] Peterson, A. W.; Heaton, R. J.; Georgiadis, R. M. The effect of surface probe density onDNA hybridization. Nucleic Acids Res. 2001, 29, 5163–5168.[16] Casanova-Moreno, J.; Yu, Z. L.; Massey-Allard, J.; Ditchburn, B.; Young, J. F.; Biz-zotto, D. Luminescence in Electrochemistry ; Springer International Publishing: Cham,2017; Chapter 2, pp 21–77.[17] Massa, W. Crystal Structure Determination, 2nd ed.; Springer-Verlag Berlin Heidelberg,2004; pp 20–22.[18] Hove, M. V.; Somorjai, G. A new microfacet notation for high-Miller-index surfaces ofcubic materials with terrace, step and kink structures. Surface Science 1980, 92, 489–518.[19] Yu, Z. L.; Casanova-Moreno, J.; Guryanov, I.; Maran, F.; Bizzotto, D. Influence of SurfaceStructure on Single or Mixed Component Self-Assembled Monolayers via in Situ Spec-troelectrochemical Fluorescence Imaging of the Complete Stereographic Triangle on aSingle Crystal Au Bead Electrode. J. Am. Chem. Soc. 2015, 137, 276–288.[20] Hamelin, A. Cyclic voltammetry at gold single-crystal surfaces. Part 1. Behaviour at low-index faces. Journal of Electroanalytical Chemistry 1996, 407, 1–11.160References[21] Hamelin, A.; Martins, A. Cyclic voltammetry at gold single-crystal surfaces. Part 2. Be-haviour of high-index faces. Journal of Electroanalytical Chemistry 1996, 407, 13–21.[22] Poirier, G. E. Mechanism of Formation of Au Vacancy Islands in Alkanethiol Monolayerson Au(111). Langmuir 1997, 13, 2019–2026.[23] Smoluchowski, R. Anisotropy of the Electronic Work Function of Metals. Phys. Rev.1941, 60, 661–674.[24] Hansson, G. V.; Flodström, S. A. Photoemission study of the bulk and surface electronicstructure of single crystals of gold. Phys. Rev. B 1978, 18, 1572–1585.[25] Lecoeur, J.; Bellier, J.; Koehler, C. Comparison of crystalographic anisotropy effects onpotential of zero charge and electronic work function for gold 111, 311, 110 and 210orientations. Electrochimica Acta 1990, 35, 1383–1392.[26] Smith, S. R.; Han, S.; McDonald, A.; Zhe, W.; Shepherd, J. L. An electrochemical ap-proach to fabricate a heterogeneous mixed monolayer on planar polycrystalline Au andits characterization with Lateral Force Microscopy. J. Electroanal. Chem. 2012, 666, 76–84.[27] Mader, S.; Feder, R.; Chaudhari, P. Recrystallization of (001) oriented gold films into(111) orientation. Thin Solid Films 1972, 14, 63 – 78.[28] Kang, S.; Bernstein, I.; Bauer, C. In-situ observations of the recrystallization process insingle crystal thin films of gold. Scripta Metallurgica 1976, 10, 693 – 696.[29] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; Chapter 1.[30] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; Chapter 13.[31] Grahame, D. C. The Electrical Double Layer and the Theory of Electrocapillarity. Chem-ical Reviews 1947, 41, 441–501.161References[32] Frumkin, A. N.; Petrii, O. A.; Damaskin, B. B. In Comprehensive Treatise of Electrochem-istry: The Double Layer ; Bockris, J. O., Conway, B. E., Yeager, E., Eds.; Springer US:Boston, MA, 1980; pp 221–289.[33] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; pp 544–557.[34] Magnussen, O. M.; Ocko, B. M.; Adzic, R. R.; Wang, J. X. X-ray diffraction studies ofordered chloride and bromide monolayers at the Au(111)-solution interface. Phys. Rev.B 1995, 51, 5510–5513.[35] de Levie, R. A simple empirical correlation between the potential of zero charge and thedensity of broken bonds. Journal of Electroanalytical Chemistry and Interfacial Electro-chemistry 1990, 280, 179–183.[36] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; p 97.[37] Ge, B.; Huang, Y.-C.; Sen, D.; Yu, H.-Z. Electrochemical investigation of DNA-modifiedsurfaces: From quantitation methods to experimental conditions. Journal of Electroana-lytical Chemistry 2007, 602, 156–162.[38] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; p 590.[39] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; p 592.[40] Bard, A. J. F. L. R. Electrochemical Method: Fundamentals and Applications; John Wiley{& Sons, INC., 2001; p 240.[41] Krysinski, P.; Chamberlain, R. V.; Majda, M. Partial Electron Transfer in OctadecanethiolBinding to Gold. Langmuir 1994, 10, 4286–4294.[42] Dilimon, V. S.; Rajalingam, S.; Delhalle, J.; Mekhalif, Z. Self-assembly mechanism of162Referencesthiol, dithiol, dithiocarboxylic acid, disulfide and diselenide on gold: an electrochemicalimpedance study. Phys. Chem. Chem. Phys. 2013, 15, 16648–16656.[43] Cooper, E.; Leggett, G. J. Static Secondary Ion Mass Spectrometry Studies of Self-Assembled Monolayers: Influence of Adsorbate Chain Length and Terminal FunctionalGroup on Rates of Photooxidation of Alkanethiols on Gold. Langmuir 1998, 14, 4795–4801.[44] Widrig, C. A.; Chung, C.; Porter, M. D. The electrochemical desorption of n-alkanethiolmonolayers from polycrystalline Au and Ag electrodes. J. Electroanal. Chem. 1991, 310,335–359.[45] Hutt, D. A.; Leggett, G. J. Influence of Adsorbate Ordering on Rates of UV Photooxida-tion of Self-Assembled Monolayers. J. Phys. Chem. 1996, 100, 6657–6662.[46] Poirier, G. E.; Pylant, E. D. The Self-Assembly Mechanism of Alkanethiols on Au(111).Science 1996, 272, 1145–1148.[47] Schoenenberger, C.; Sondag-Huethorst, J. A. M.; Jorritsma, J.; Fokkink, L. G. J. WhatAre the "Holes" in Self-Assembled Monolayers of Alkanethiols on Gold? Langmuir 1994,10, 611–614.[48] Qian, Y.; Yang, G.; Yu, J.; Jung, T. A.; Liu, G.-y. Structures of Annealed DecanethiolSelf-Assembled Monolayers on Au(111):an Ultrahigh Vacuum Scanning Tunneling Mi-croscopy Study. Langmuir 2003, 19, 6056–6065.[49] Toerker, M.; Staub, R.; Fritz, T.; Schmitz-HÃŒbsch, T.; Sellam, F.; Leo, K. Annealeddecanethiol monolayers on Au(111): intermediate phases between structures with highand low molecular surface density. Surface Science 2000, 445, 100–108.[50] Shon, Y.-S.; Lee, T. R. A Steady-State Kinetic Model Can Be Used to Describe theGrowth of Self-Assembled Monolayers (SAMs) on Gold. J. Phys. Chem. B 2000, 104,8182–8191.163References[51] Bensebaa, F.; Voicu, R.; Huron, L.; Ellis, T. H.; Kruus, E. Kinetics of Formation of Long-Chain n-Alkanethiolate Monolayers on Polycrystalline Gold. Langmuir 1997, 13, 5335–5340.[52] Yamada, R.; Uosaki, K. In Situ Scanning Tunneling Microscopy Observation of the Self-Assembly Process of Alkanethiols on Gold(111) in Solution. Langmuir 1998, 14, 855–861.[53] Wu, H.; Sotthewes, K.; Kumar, A.; Vancso, G. J.; Schön, P. M.; Zandvliet, H. J. W.Dynamics of Decanethiol Self-Assembled Monolayers on Au(111) Studied by Time-Resolved Scanning Tunneling Microscopy. Langmuir 2013, 29, 2250–2257.[54] Chidsey, C. E. D.; Bertozzi, C. R.; Putvinski, T. M.; Mujsce, A. M. Coadsorptionof ferrocene-terminated and unsubstituted alkanethiols on gold: electroactive self-assembled monolayers. J. Am. Chem. Soc. 1990, 112, 4301–4306.[55] Collard, D. M.; Fox, M. A. Use of electroactive thiols to study the formation and exchangeof alkanethiol monolayers on gold. Langmuir 1991, 7, 1192–1197.[56] Kakiuchi, T.; Sato, K.; Iida, M.; Hobara, D.; Imabayashi, S.; Niki, K. Phase Separation ofAlkanethiol Self-Assembled Monolayers during the Replacement of Adsorbed Thiolateson Au(111) with Thiols in Solution. Langmuir 2000, 16, 7238–7244.[57] Chung, C.; Lee, M. Exchange of self-assembled thiol monolayers on gold: characteriza-tion by FT-IR external reflection spectroscopy. J. Electroanal. Chem. 1999, 468, 91–97.[58] Camillone, N.; Chidsey, C. E. D.; Liu, G. Y.; Scoles, G. Substrate dependence of thesurface structure and chain packing of docosyl mercaptan self-assembled on the (111),(110), and (100) faces of single crystal gold. The Journal of Chemical Physics 1993, 98,4234–4245.[59] Camillone, N.; Leung, T.; Scoles, G. A low energy helium atom diffraction study of de-canethiol self-assembled on Au(111). Surf. Sci. 1997, 373, 333–349.164References[60] Tang, L.; Li, F.; Zhou, W.; Guo, Q. The structure of methylthiolate and ethylthiolate mono-layers on Au(111): Absence of the (√3 ×√3)R30◦ phase. Surface Science 2012, 606,L31–L35.[61] Pensa, E.; Cortes, E.; Corthey, G.; Carro, P.; Vericat, C.; Fonticelli, M. H.; Benitez, G.;Rubert, A. A.; Salvarezza, R. C. The Chemistry of the Sulfur-Gold Interface: In Searchof a Unified Model. Acc. Chem. Res. 2012, 45, 1183–1192.[62] Voznyy, O.; Dubowski, J. J.; Yates, J. T.; Maksymovych, P. The Role of Gold Adatoms andStereochemistry in Self-Assembly of Methylthiolate on Au(111). Journal of the AmericanChemical Society 2009, 131, 12989–12993.[63] Noh, J.; Kato, H. S.; Kawai, M.; Hara, M. Surface Structure and Interface Dynamics ofAlkanethiol Self-Assembled Monolayers on Au(111). The Journal of Physical ChemistryB 2006, 110, 2793–2797.[64] Shnidman, Y.; Ulman, A.; Eilers, J. E. Effect of perturbing strata on chain conformationsand ordering in closely packed layered structures of chain molecules. Langmuir 1993,9, 1071–1081.[65] Grumelli, D.; Cristina, L. J.; Maza, F. L.; Carro, P.; Ferron, J.; Kern, K.; Salvarezza, R. C.Thiol Adsorption on the Au(100)-hex and Au(100)-(1× 1) Surfaces. The Journal of Phys-ical Chemistry C 2015, 119, 14248–14254.[66] Vericat, C.; Vela, M. E.; Salvarezza, R. C. Self-assembled monolayers of alkanethiols onAu(111): surface structures, defects and dynamics. Phys. Chem. Chem. Phys. 2005, 7,3258–3268.[67] Sabatani, E.; Rubinstein, I. Organized self-assembling monolayers on electrodes. 2.Monolayer-based ultramicroelectrodes for the study of very rapid electrode kinetics. TheJournal of Physical Chemistry 1987, 91, 6663–6669.[68] Finklea, H. O.; Avery, S.; Lynch, M.; Furtsch, T. Blocking oriented monolayers of alkylmercaptans on gold electrodes. Langmuir 1987, 3, 409–413.165References[69] Ramírez, P.; Andreu, R.; Cuesta, A.; Calzado, C. J.; Calvente, J. J. Determination of thePotential of Zero Charge of Au(111) Modified with Thiol Monolayers. Anal. Chem. 2007,79, 6473–6479.[70] Lemay, D. M.; Shepherd, J. L. Electrochemical fabrication of a heterogeneous binarySAM on polycrystalline Au. Electrochimica Acta 2008, 54, 388–393.[71] Doneux, T.; Steichen, M.; Rache, A. D.; Buess-Herman, C. Influence of the crystallo-graphic orientation on the reductive desorption of self-assembled monolayers on goldelectrodes. J. Electroanal. Chem. 2010, 649, 164–170.[72] Musgrove, A.; Kell, A.; Bizzotto, D. Fluorescence Imaging of the Oxidative Desorption ofa BODIPY-Alkyl-Thiol Monolayer Coated Au Bead. Langmuir 2008, 24, 7881–7888.[73] Shepherd, J. L.; Kell, A.; Chung, E.; Sinclar, C. W.; Workentin, M. S.; Bizzotto, D. Se-lective Reductive Desorption of a SAM-Coated Gold Electrode Revealed Using Fluores-cence Microscopy. J. Am. Chem. Soc. 2004, 126, 8329–8335.[74] Skoog, D. A.; Holler, F. J.; Crouch, S. R. Principles of instrumental analysis, 6th ed.;Belmont, CA : Thomson, Brooks/Cole, 2007; p 194.[75] Barnes, W. L. Fluorescence near interfaces: The role of photonic mode density. J. Mod.Opt. 1998, 45, 661–699.[76] Chance, R. R.; Prock, A.; Silbey, R. Advances in Chemical Physics; John Wiley & Sons,Inc., 2007; pp 1–65.[77] Yeung, M. S.; Gustafson, T. K. Spontaneous emission near an absorbing dielectric sur-face. Phys. Rev. A 1996, 54, 5227–5242.[78] Franklin, R. E.; Gosling, R. G. Molecular Configuration in Sodium Thymonucleate. Na-ture 1953, 171, 740–741.[79] Watson, J. D.; Crick, F. H. C. Molecular Structure of Nucleic Acids: A Structure for De-oxyribose Nucleic Acid. Nature 1953, 171, 737–738.[80] Dixon, B. Unfolding the DNA story. Current Biology 2003, 13, R295 – R296.166References[81] Tibanyenda, N.; De Bruin, S. H.; Haasnoot, C. A. G.; Van Der Marel, G. A.;Van Boom, J. H.; Hilbers, C. W. The effect of single base-pair mismatches on the du-plex stability of d(T-A-T-T-A-A-T-A-T-C-A-A-G-T-T-G) . d(C-A-A-C-T-T-G-A-T-A-T-T-A-A-T-A). European Journal of Biochemistry 1984, 139, 19–27.[82] Manning, G. S. Limiting Laws and Counterion Condensation in Polyelectrolyte SolutionsI. Colligative Properties. The Journal of Chemical Physics 1969, 51, 924–933.[83] Tan, Z.-J.; Chen, S.-J. Nucleic Acid Helix Stability: Effects of Salt Concentration, CationValence and Size, and Chain Length. Biophysical Journal 2006, 90, 1175–1190.[84] Williams, L. D.; Maher III, L. J. Electrostatic Mechanisms of DNA Deformation. AnnualReview of Biophysics and Biomolecular Structure 2000, 29, 497–521.[85] Gebala, M.; Schuhmann, W. Understanding properties of electrified interfaces as a pre-requisite for label-free DNA hybridization detection. Phys. Chem. Chem. Phys. 2012, 14,14933–14942.[86] Tinland, B.; Pluen, A.; Sturm, J.; Weill, G. Persistence Length of Single-Stranded DNA.Macromolecules 1997, 30, 5763–5765.[87] Bustamante, C.; Bryant, Z.; Smith, S. B. Ten years of tension: single-molecule DNAmechanics. Nature 2003, 421, 423.[88] Kaiser, W.; Rant, U. Conformations of End-Tethered DNA Molecules on Gold Surfaces:Influences of Applied Electric Potential, Electrolyte Screening, and Temperature. Journalof the American Chemical Society 2010, 132, 7935–7945.[89] Herne, T. M.; Tarlov, M. J. Characterization of DNA Probes Immobilized on Gold Sur-faces. J. Am. Chem. Soc. 1997, 119, 8916–8920.[90] Berg, J.; Tymoczko, J.; Stryer, L. Biochemistry, Fifth Edition; W.H. Freeman, 2002.[91] Levicky, R.; Herne, T. M.; Tarlov, M. J.; Satija, S. K. Using Self-Assembly To Control theStructure of DNA Monolayers on Gold: A Neutron Reflectivity Study. J. Am. Chem. Soc.1998, 120, 9787–9792.167References[92] Li, Z.; Niu, T.; Zhang, Z.; Chen, R.; Feng, G.; Bi, S. Exploration of the specific struc-tural characteristics of thiol-modified single-stranded DNA self-assembled monolayerson gold by a simple model. Biosens. Bioelectron. 2011, 26, 4564–4570.[93] Wong, K.-Y.; Montgomery Pettitt, B. A study of DNA tethered to a surface by an all-atommolecular dynamics simulation. Theor. Chem. Acc. 2001, 106, 233–235.[94] Yao, L.; Sullivan, J.; Hower, J.; He, Y.; Jiang, S. Packing structures of single-strandedDNA and double-stranded DNA thiolates on Au(111): A molecular simulation study. TheJournal of Chemical Physics 2007, 127, 195101.[95] Lee, O.-S.; Schatz, G. C. Interaction between DNAs on a Gold Surface. The Journal ofPhysical Chemistry C 2009, 113, 15941–15947.[96] Petrovykh, D. Y.; Kimura-Suda, H.; Whitman, L. J.; Tarlov, M. J. Quantitative Analysisand Characterization of DNA Immobilized on Gold. J. Am. Chem. Soc. 2003, 125, 5219–5226.[97] Ambia-Garrido, J.; Vainrub, A.; Pettitt, B. M. A model for Structure and Thermodynam-ics of ssDNA and dsDNA Near a Surface: a Coarse Grained Approach. Comput PhysCommun 2010, 181, 2001–2007.[98] Erts, D.; Polyakov, B.; Olin, H.; Tuite, E. Spatial and Mechanical Properties of Dilute DNAMonolayers on Gold Imaged by AFM. The Journal of Physical Chemistry B 2003, 107,3591–3597.[99] Josephs, E. A.; Ye, T. Nanoscale Spatial Distribution of Thiolated DNA on Model NucleicAcid Sensor Surfaces. ACS Nano 2013, 7, 3653–3660.[100] Georgiadis, R.; Peterlinz, K. P.; Peterson, A. W. Quantitative measurements and mod-eling of kinetics in nucleic acid monolayer films using SPR spectroscopy. J. Am. Chem.Soc. 2000, 122, 3166–3173.[101] Wackerbarth, H.; Grubb, M.; Zhang, J.; Hansen, A. G.; Ulstrup, J. Long-Range Orderof Organized Oligonucleotide Monolayers on Au(111) Electrodes. Langmuir 2004, 20,1647–1655.168References[102] Wolf, L. K.; Gao, Y.; Georgiadis, R. M. Sequence-dependent DNA immobilization: spe-cific versus nonspecific contributions. Langmuir 2004, 20, 3357–3361.[103] Murphy, J. N.; Cheng, A. K. H.; Yu, H.-Z.; Bizzotto, D. On the Nature of DNA Self-Assembled Monolayers on Au: Measuring Surface Heterogeneity with Electrochemicalin Situ Fluorescence Microscopy. J. Am. Chem. Soc. 2009, 131, 4042–4050.[104] Keighley, S. D.; Li, P.; Estrela, P.; Migliorato, P. Optimization of DNA immobilizationon gold electrodes for label-free detection by electrochemical impedance spectroscopy.Biosens. Bioelectron. 2008, 23, 1291–1297.[105] Boozer, C.; Chen, S.; Jiang, S. Controlling DNA Orientation on Mixed ssDNA/OEGSAMs. Langmuir 2006, 22, 4694–4698.[106] Doneux, T.; De Rache, A.; Triffaux, E.; Meunier, A.; Steichen, M.; Buess-Herman, C.Optimization of the Probe Coverage in DNA Biosensors by a One-Step CoadsorptionProcedure. ChemElectroChem 2014, 1, 147–157.[107] Rao, A. N.; Grainger, D. W. Biophysical Properties Of Nucleic Acids At Surfaces RelevantTo Microarray Performance. Biomater Sci 2014, 2, 436–471.[108] Cantor, R. C. Genomics. The Science and Technology Behind the Human GenomeProject ; 1999.[109] Wong, E. L. S.; Chow, E.; Gooding, J. J. DNA Recognition Interfaces: The Influence ofInterfacial Design on the Efficiency and Kinetics of Hybridization. Langmuir 2005, 21,6957–6965.[110] Tyagi, S.; Kramer, F. R. Molecular Beacons: Probes that Fluoresce upon Hybridization.Nat. Biotechnol. 1996, 14, 303–308.[111] Du, H.; Disney, M. D.; Miller, B. L.; Krauss, T. D. Hybridization-Based Unquenchingof DNA Hairpins on Au Surfaces: Prototypical "Molecular Beacon" Biosensors. J. Am.Chem. Soc. 2003, 125, 4012–4013.169References[112] Fan, C.; Plaxco, K. W.; Heeger, A. J. Electrochemical interrogation of conformationalchanges as a reagentless method for the sequence-specific detection of DNA. Proceed-ings of the National Academy of Sciences 2003, 100, 9134–9137.[113] Xiao, Y.; Lubin, A. A.; Heeger, A. J.; Plaxco, K. W. Label-Free Electronic Detection ofThrombin in Blood Serum by Using an Aptamer-Based Sensor. Angewandte ChemieInternational Edition 2005, 44, 5456–5459.[114] Xiao, Y.; Lubin, A. A.; Baker, B. R.; Plaxco, K. W.; Heeger, A. J. Single-step electronicdetection of femtomolar DNA by target-induced strand displacement in an electrode-bound duplex. Proceedings of the National Academy of Sciences 2006, 103, 16677–16680.[115] Yu, H.-Z.; Luo, C.-Y.; Sankar, C. G.; Sen, D. Voltammetric Procedure for ExaminingDNA-Modified Surfaces: Quantitation, Cationic Binding Activity, and Electron-TransferKinetics. Anal. Chem. 2003, 75, 3902–3907.[116] Rant, U.; Arinaga, K.; Scherer, S.; Pringsheim, E.; Fujita, S.; Yokoyama, N.; Tornow, M.;Abstreiter, G. Switchable DNA interfaces for the highly sensitive detection of label-freeDNA targets. Proceedings of the National Academy of Sciences 2007, 104, 17364–17369.[117] Rant, U.; Pringsheim, E.; Kaiser, W.; Arinaga, K.; Knezevic, J.; Tornow, M.; Fujita, S.;Yokoyama, N.; Abstreiter, G. Detection and Size Analysis of Proteins with SwitchableDNA Layers. Nano Lett. 2009, 9, 1290–1295.[118] Arinaga, K.; Rant, U.; Knezevicb, J.; Pringsheim, E.; Tornow, M.; Fujita, S.; Abstreiter, G.;Yokoyama, N. Controlling the surface density of DNA on gold by electrically induceddesorption. Biosens. Bioelectron. 2007, 23, 326–331.[119] Thorp, H. H. Reagentless detection of DNA sequences on chemically modified elec-trodes. Trends Biotechnol. 2003, 21, 522–524.[120] Steel, A. B.; Herne, T. M.; Tarlov, M. J. Electrochemical Quantitation of DNA Immobilizedon Gold. Anal. Chem. 1998, 70, 4670–4677.170References[121] Ho, P. S.; Frederick, C. A.; Saal, D.; Wang, A. H.-J.; Rich, A. The Interactions of Ruthe-nium Hexaammine with Z-DNA: Crystal Structure of a Ru(NH3)+3 6 Salt of d(CGCGCG)at 1.2 A Resolution. J. Biomol. Struct. Dyn. 1987, 4, 521–534.[122] Thomas, T.; Messner, R. P. Hexammineruthenium (III) chloride: a highly efficient pro-moter of the B-DNA to Z-DNA transition of poly-(dG-m5dC)A poly(dG-m5dC) andpoly(dG-dC)A poly(dG-dC). Biochimie 1988, 70, 221–226.[123] Steichen, M.; Doneux, T.; Buess-Herman, C. On the adsorption of hexaammineruthe-nium (III) at anionic self-assembled monolayers. Electrochimica Acta 2008, 53, 6202 –6208.[124] Turcu, F.; Schulte, A.; Hartwich, G.; Schuhmann, W. Label-Free Electrochemical Recog-nition of DNA Hybridization by Means of Modulation of the Feedback Current in SECM.Angew. Chem 2004, 43, 3482–3485.[125] Li, D.; Zou, X.; Shen, Q.; Dong, S. Kinetic study of DNA/DNA hybridization with electro-chemical impedance spectroscopy. Electrochem. Commun. 2007, 9, 191–196.[126] Lam, B.; Das, J.; Holmes, R. D.; Live, L.; Sage, A.; Sargent, E. H.; Kelley, S. O. Solution-based circuits enable rapid and multiplexed pathogen detection. Nature Communica-tions 2013, 4, 2001.[127] Kelley, S. O.; Boon, E. M.; Barton, J. K.; Jackson, N. M.; Hill, M. G. a. Single-basemismatch detection based on charge transduction through DNA. Nucleic Acids Res.1999, 27 (24), 4830â4837.[128] Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Hill, M. G. Electrochemistry of MethyleneBlue Bound to a DNA-Modified Electrode. Bioconjug. Chem. 1997, 8, 31–37.[129] Boon, E. M.; Ceres, D. M.; Drummond, T. G.; Hill, M. G.; Barton, J. K. Mutation detectionby electrocatalysis at DNA-modified electrodes. Nat. Biotechnol. 2000, 18, 1096–1100.[130] Ferapontova, E. E. Hybridization Biosensors Relying on Electrical Properties of NucleicAcids. Electroanalysis 2017, 29, 6–13.171References[131] Kang, D.; Zuo, X.; Yang, R.; Xia, F.; Plaxco, K. W.; White, R. J. Comparing the Propertiesof Electrochemical-Based DNA Sensors Employing Different Redox Tags. Anal. Chem.2009, 81, 9109–9113.[132] Vallée-Bélisle, A.; Ricci, F.; Plaxco, K. W. Thermodynamic basis for the optimization ofbinding-induced biomolecular switches and structure-switching biosensors. Proceedingsof the National Academy of Sciences 2009, 106, 13802–13807.[133] Deng, B.; Lin, Y.; Wang, C.; Li, F.; Wang, Z.; Zhang, H.; Li, X.-F.; Le, X. C. Aptamerbinding assays for proteins: The thrombin example: A review. Anal. Chim. Acta 2014,837, 1–15.[134] Ricci, F.; Lai, R. Y.; Heeger, A. J.; Plaxco, K. W.; Sumner, J. J. Effect of MolecularCrowding on the Response of an Electrochemical DNA Sensor. Langmuir 2007, 23,6827–6834.[135] Li, X.; Quigg, R. J.; Zhou, J.; Gu, W.; Nagesh Rao, P.; Reed, E. F. Clinical utility of mi-croarrays: current status, existing challenges and future outlook. Curr Genomics 2008,9, 466–474.[136] Zhang, J.; Lao, R.; Song, S.; Yan, Z.; Fan, C. Design of an Oligonucleotide-IncorporatedNonfouling Surface and Its Application in Electrochemical DNA Sensors for Highly Sen-sitive and Sequence-Specific Detection of Target DNA. Anal. Chem. 2008, 80, 9029–9033.[137] Lubin, A. A.; Plaxco, K. W. Folding-Based Electrochemical Biosensors: The Case forResponsive Nucleic Acid Architectures. Acc. Chem. Res. 2010, 43, 496–505.[138] Gorshkova, I. I.; Svitel, J.; Razjouyan, F.; Schuck, P. Bayesian Analysis of Heterogeneityin the Distribution of Binding Properties of Immobilized Surface Sites. Langmuir 2008,24, 11577–11586.[139] Lee, C.-Y.; Canavan, H. E.; Gamble, L. J.; Castner, D. G. Evidence of Impurities inThiolated Single-Stranded DNA Oligomers and Their Effect on DNA Self-Assembly onGold. Langmuir 2005, 21, 5134–5141.172References[140] Yu, Z. L.; Yang, C. W. T.; Triffaux, E.; Doneux, T.; Turner, R. F. B.; Bizzotto, D. Mea-suring and Remediating Nonspecific Modifications of Gold Surfaces Using a Coupledin Situ Electrochemical Fluorescence Microscopic Methodology. Analytical Chemistry2017, 89, 886–894.[141] Lao, R.; Song, S.; Wu, H.; Wang, L.; Zhang, Z.; He, L.; Fan, C. Electrochemical Interro-gation of DNA Monolayers on Gold Surfaces. Anal. Chem. 2005, 77, 6475–6480.[142] Steel, A.; Levicky, R.; Herne, T.; Tarlov, M. Immobilization of Nucleic Acids at SolidSurfaces: Effect of Oligonucleotide Length on Layer Assembly. Biophys. J. 2000, 79,975–981.[143] Gong, P.; Levicky, R. DNA surface hybridization regimes. Proceedings of the NationalAcademy of Sciences 2008, 105, 5301–5306.[144] White, R. J.; Phares, N.; Lubin, A. A.; Xiao, Y.; Plaxco, K. W. Optimization of Electro-chemical Aptamer-Based Sensors via Optimization of Probe Packing Density and Sur-face Chemistry. Langmuir 2008, 24, 10513–10518.[145] Doudevski, I.; Schwartz, D. K. Self-assembled monolayers in the context of epitaxial filmgrowth. Appl. Surf. Sci. 2001, 175-176, 17–26, 10th International Conference on SolidFilms and Surfaces.[146] Xiao, Y.; Lai, R. Y.; Plaxco, K. W. Preparation of electrode-immobilized, redox-modifiedoligonucleotides for electrochemical DNA and aptamer-based sensing. Nat. Protoc.2007, 2, 2875–2880.[147] Jambrec, D.; Gebala, M.; La Mantia, F., F.Mantia; Schuhmann, W. Potential-AssistedDNA Immobilization as a Prerequisite for Fast and Controlled Formation of DNA Mono-layers. Angewandte Chemie International Edition 2015, 54, 15064–15068.[148] Haag, A.-L.; Toader, V.; Lennox, R. B.; Grutter, P. Selective in situ potential-assistedSAM formation on multi electrode arrays. Nanotechnology 2016, 27, 455501.173References[149] Collman, J. P.; Hosseini, A.; Eberspacher, T. A.; Chidsey, C. E. D. Selective AnodicDesorption for Assembly of Different Thiol Monolayers on the Individual Electrodes of anArray. Langmuir 2009, 25, 6517–6521.[150] Zhong, C.-J.; Woods, N. T.; Dawson, G.; Porter, M. D. Formation of thiol-based mono-layers on gold: implications from open circuit potential measurements. ElectrochemistryCommunications 1999, 1, 17–21.[151] Ron, H.; Rubinstein, I. Self-Assembled Monolayers on Oxidized Metals. 3. Alkylthiol andDialkyl Disulfide Assembly on Gold under Electrochemical Conditions. J. Am. Chem.Soc. 1998, 120, 13444–13452.[152] Paik, W.-k.; Eu, S.; Lee, K.; Chon, S.; Kim, M. Electrochemical Reactions in Adsorp-tion of Organosulfur Molecules on Gold and Silver: Potential Dependent Adsorption.Langmuir 2000, 16, 10198–10205.[153] Kroener, F.; Heerwig, A.; Kaiser, W.; Mertig, M.; Rant, U. Electrical Actuation of a DNAOrigami Nanolever on an Electrode. Journal of the American Chemical Society 2017,139, 16510–16513.[154] Bizzotto, D.; Pettinger, B. Fluorescence Imaging Studies of the Electrochemical Adsorp-tion/Desorption of Octadecanol. Langmuir 1999, 15, 8309–8314.[155] Casanova-Moreno, J. R.; Bizzotto, D. What Happens to the Thiolates Created by Reduc-tively Desorbing SAMs? An in Situ Study Using Fluorescence Microscopy and Electro-chemistry. Langmuir 2013, 29, 2065–2074.[156] Rant, U.; Arinaga, K.; Fujita, S.; Yokoyama, N.; Abstreiter, G.; Tornow, M. DynamicElectrical Switching of DNA Layers on a Metal Surface. Nano Lett. 2004, 4, 2441–2445.[157] Rant, U.; Arinaga, K.; Fujita, S.; Yokoyama, N.; Abstreiter, G.; Tornow, M. StructuralProperties of Oligonucleotide Monolayers on Gold Surfaces Probed by FluorescenceInvestigations. Langmuir 2004, 20, 10086–10092.174References[158] Meunier, A.; Triffaux, E.; Bizzotto, D.; Buess-Herman, C.; Doneux, T. InâSitu Fluores-cence Microscopy Study of the Interfacial Inhomogeneity of DNA Mixed Self-AssembledMonolayers at Gold Electrodes. ChemElectroChem 2015, 2, 434–442.[159] Gu, Q.; Nanney, W.; Cao, H. H.; Wang, H.; Ye, T. Single Molecule Profiling of Molecu-lar Recognition at a Model Electrochemical Biosensor. J. Am. Chem. Soc. 2018, 140,14134–14143.[160] Grubb, M.; Wackerbarth, H.; Ulstrup, J. Identification of Single-Strand DNA by in situScanning Tunneling Microscopy. J. Am. Chem. Soc. 2006, 128, 7734–7735.[161] Grubb, M.; Wackerbarth, H.; Wengel, J.; Ulstrup, J. Direct Imaging of Hexaamine-Ruthenium(III) in Domain Boundaries in Monolayers of Single-Stranded DNA. Langmuir2007, 23, 1410–1413.[162] Casanova-Moreno, J. R.; Bizzotto, D. Frequency response analysis of potential-modulated orientation changes of a DNA self assembled layer using spatially resolvedfluorescence measurements. Electrochim. Acta 2015, 162, 62–71.[163] Trasatti, S.; Petrii, O. Real surface area measurements in electrochemistry. Journal ofElectroanalytical Chemistry 1992, 327, 353 – 376, An International Journal Devoted toall Aspects of Electrode Kinetics, Interfacial Structure, Properties of Electrolytes, Colloidand Biological Electrochemistry.[164] Zuker, M. Mfold web server for nucleic acid folding and hybridization prediction. NucleicAcids Res. 2003, 31, 3406–3415.[165] Kibbe, W. A. OligoCalc: an online oligonucleotide properties calculator. Nucleic AcidsRes. 2007, 35, W43–46.[166] Shepherd, J. L.; Bizzotto, D. Characterization of Mixed Alcohol Monolayers Adsorbedonto a Au(111) Electrode Using Electro-fluorescence Microscopy. Langmuir 2006, 22,4869–4876.[167] Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W. NIH Image to ImageJ: 25 years of imageanalysis. Nature Methods 2012, 9, 671.175References[168] Leung, K. K.; Gaxiola, A. D.; Yu, H.-Z.; Bizzotto, D. Tailoring the DNA SAM surface den-sity on different surface crystallographic features using potential assisted thiol exchange.Electrochim. Acta 2018, 261, 188–197.[169] Leung, K. K.; Yu, H.-Z.; Bizzotto, D. Electrodepositing DNA Self-Assembled Monolayerson Au: Detailing the Influence of Electrical Potential Perturbation and Surface Crystal-lography. ACS Sensors 2019, 4, 513–520.[170] Sahalov, H.; O’Brien, B.; Stebe, K. J.; Hristova, K.; Searson, P. C. Influence of AppliedPotential on the Impedance of Alkanethiol SAMs. Langmuir 2007, 23, 9681–9685.[171] Boubour, E.; Lennox, R. B. Stability of ω-Functionalized Self-Assembled Monolayers asa Function of Applied Potential. Langmuir 2000, 16, 7464–7470.[172] Sondag-Huethorst, J. A. M.; Fokkink, L. G. J. Electrochemical Characterization of Func-tionalized Alkanethiol Monolayers on Gold. Langmuir 1995, 11, 2237–2241.[173] Boubour, E.; Lennox, R. B. Potential-induced defects in n-alkanethiol self-assembledmonolayers monitored by impedance spectroscopy. The Journal of Physical ChemistryB 2000, 104, 9004–9010.[174] Everett, W. R.; Welch, T. L.; Reed, L.; Fritsch-Faules, I. Potential-Dependent Stabilityof Self-Assembled Organothiols on Gold Electrodes in Methylene Chloride. AnalyticalChemistry 1995, 67, 292–298.[175] Diao, P.; Hou, Q.; Guo, M.; Xiang, M.; Zhang, Q. Effect of substrate potentials on thestructural disorders of alkanethiol monolayers prepared by electrochemically directedassembly. J. Electroanal. Chem. 2006, 597, 103–110.[176] Darwish, N.; Eggers, P. K.; Ciampi, S.; Zhang, Y.; Tong, Y.; Ye, S.; Paddon-Row, M. N.;Gooding, J. J. Reversible potential-induced structural changes of alkanethiol monolayerson gold surfaces. Electrochem. Commun. 2011, 13, 387–390.[177] Anderson, M. R.; Gatin, M. Effect of applied potential upon the in situ structure of self-assembled monolayers on gold electrodes. Langmuir 1994, 10, 1638–1641.176References[178] Schweizer, M.; Manolova, M.; Kolb, D. Potential-induced structure transitions in self-assembled monolayers: II. Propanethiol on Au(100). Surface Science 2008, 602, 3303– 3307.[179] Schweizer, M.; Hagenström, H.; Kolb, D. Potential-induced structure transitions in self-assembled monolayers: ethanethiol on Au(100). Surface Science 2001, 490, L627 –L636.[180] Hagenström, H.; Schneeweiss, M. A.; Kolb, D. M. Modification of a Au(111) Electrodewith Ethanethiol. 1. Adlayer Structure and Electrochemistry. Langmuir 1999, 15, 2435–2443.[181] Abbott, N. L.; Gorman, C. B.; Whitesides, G. M. Active Control of Wetting Using AppliedElectrical Potentials and Self- Assembled Monolayers. Langmuir 1995, 11, 16–18.[182] Fisher, E. A.; Leung, K. K.; Casanova-Moreno, J.; Masuda, T.; Young, J. F.; Bizzotto, D.Quantifying the selective modification of Au(111) facets via electrochemical and elec-troless treatments for manipulating gold nanorod surface composition. Langmuir 2017,acs.langmuir.7b03021.[183] Cherstvy, A. G. Electrostatic interactions in biological DNA-related systems. Phys.Chem. Chem. Phys. 2011, 13, 9942–9968.[184] Doneux, T.; Nichols, R. J.; Buess-Herman, C. Dissolution kinetics of octadecanethiolatemonolayers electro-adsorbed on Au(111). J. Electroanal. Chem. 2008, 621, 267–276.[185] Hobara, D.; Miyake, K.; Imabayashi, S.-i.; Niki, K.; Kakiuchi, T. In-Situ Scanning Tunnel-ing Microscopy Imaging of the Reductive Desorption Process of Alkanethiols on Au(111).Langmuir 1998, 14, 3590–3596.[186] Yang, G.; Amro, N. A.; Starkewolfe, Z. B.; Liu, G.-y. Molecular-Level Approach To InhibitDegradations of Alkanethiol Self-Assembled Monolayers in Aqueous Media. Langmuir2004, 20, 3995–4003.[187] Kornyshev, A. A. Physics of DNA: unravelling hidden abilities encoded in the structure of’the most important molecule’. Phys. Chem. Chem. Phys. 2010, 12, 12352–12378.177References[188] Li, Z.; Niu, T.; Zhang, Z.; Feng, G.; Bi, S. Studies on the effect of solvents on self-assembly of thioctic acid and Mercaptohexanol on gold. Thin Solid Films 2011, 519,4225–4233.[189] Shi, Z.; Lipkowski, J. Chloride adsorption at the Au(111) electrode surface. J. Elec-troanal. Chem. 1996, 403, 225–239.[190] Mostafa, M.; Payam, K.; Timo, J.; Pedrero-Ernesto, P.; Margret, G.; Harald, I.; B., L. N.;Wolfgang, S. Chlorine-Enhanced Surface Mobility of Au(100). ChemPhysChem 2013,14, 233–236.[191] Ye, S.; Ishibashi, C.; Uosaki, K. Anisotropic Dissolution of an Au(111) Electrode in Per-chloric Acid Solution Containing Chloride Anion Investigated by in Situ STMThe Impor-tant Role of Adsorbed Chloride Anion. Langmuir 1999, 15, 807–812.[192] Quan, X.; Heiskanen, A.; Tenje, M.; Boisen, A. In-situ monitoring of potential enhancedDNA related processes using electrochemical quartz crystal microbalance with dissipa-tion (EQCM-D). Electrochem. Commun. 2014, 48, 111–114.[193] Jambrec, D.; Conzuelo, F.; Estrada-Vargas, A.; Schuhmann, W. Potential-Pulse-AssistedFormation of Thiol Monolayers within Minutes for Fast and Controlled Electrode SurfaceModification. ChemElectroChem 2016, 3, 1484–1489.[194] Smith, S. R.; Guerra, E.; Siemann, S.; Shepherd, J. L. Au dissolution during the anodicresponse of short-chain alkylthiols with polycrystalline Au electrodes. Electrochim. Acta2011, 56, 8291–8298.[195] Liu, Y.-F.; Lee, Y.-L.; Yang, Y.-C.; Jian, Z.-Y.; Dow, W.-P.; Yau, S.-L. Effect of Chloride Ionson the Adsorption of 3-Mercapto-1-propanesulfonic acid and Bis(3-sulfopropyl)-disulfideon a Au(111) Surface. Langmuir 2010, 26, 13263–13271.[196] Magnussen, O. M. Ordered Anion Adlayers on Metal Electrode Surfaces. Chem. Rev.2002, 102, 679–726.[197] Cuesta, A.; Kolb, D. M. The structure of bromide and chloride adlayers on Au (100)electrodes: an in situ STM study. Surf. Sci. 2000, 465, 310–316.178References[198] Ye, S.; Ishibashi, C.; Shimazu, K.; Uosaki, K. An In Situ Electrochemical Quartz CrystalMicrobalance Study of the Dissolution Process of a Gold Electrode in Perchloric AcidSolution Containing Chloride Ion. J. Electrochem. Soc. 1998, 145, 1614–1623.[199] Trevor, D. J.; Chidsey, C. E. D.; Loiacono, D. N. In Situ Scanning-Tunneling-MicroscopeObservation of Roughening, Annealing, and Dissolution of Gold (111) in an Electro-chemical Cell. Phys. Rev. Lett. 1989, 62, 929–932.[200] Honbo, H.; Sugawara, S.; Itaya, K. Detailed in-situ scanning tunneling microscopy ofsingle crystal planes of gold(111) in aqueous solutions. Analytical Chemistry 1990, 62,2424–2429.[201] Haiss, W.; Sass, J.-K. The STM imaging of an ordered (AuCl3)2 surface species duringchloride-induced dissolution of Au(111). Journal of Electroanalytical Chemistry 1997,431, 15 – 17.[202] Zamborini, F. P.; Crooks, R. M. Corrosion Passivation of Gold by n-Alkanethiol Self-Assembled Monolayers:âEffect of Chain Length and End Group. Langmuir 1998, 14,3279–3286.[203] Kyle, S.; Aggeli, A.; Ingham, E.; McPherson, M. J. Production of self-assembling bioma-terials for tissue engineering. Trends in Biotechnology 2009, 27, 423 – 433.[204] Erickson, H. P. Size and shape of protein molecules at the nanometer level determinedby sedimentation, gel filtration, and electron microscopy. Biol Proced Online 2009, 11,32–51.[205] Mahshid, S. S.; Vallée-Bélisle, A.; Kelley, S. O. Biomolecular Steric Hindrance Effects AreEnhanced on Nanostructured Microelectrodes. Analytical Chemistry 2017, 89, 9751–9757.[206] Dupont, D. M.; Thuesen, C. K.; Bøtkjær, K. A.; Behrens, M. A.; Dam, K.; Sørensen, H. P.;Pedersen, J. S.; Ploug, M.; Jensen, J. K.; Andreasen, P. A. Protein-binding RNA ap-tamers affect molecular interactions distantly from their binding sites. PLoS One 2015,10, e0119207–e0119207.179[207] Martens, I.; Fisher, E. A.; Bizzotto, D. Direct Mapping of Heterogeneous Surface Cov-erage in DNA-Functionalized Gold Surfaces with Correlated Electron and FluorescenceMicroscopy. Langmuir 2018, 34, 2425–2431.180Appendix AT-test Comparison of DNA CoveragesThis section explains how the t-tests were done to compare the fluorescence intensity mea-surements of DNA SAMs made atEdep withESWdep in Chapter 6. T-tests were performed for sam-ples with unequal variances on fluorescence intensities from DNA SAMs made at Edep=0.4Vand ESWdep =0.4V/-0.3V(50 Hz) for each of the crystallographic features. The t value was calcu-lated using equation A.1 where FDC and FSW correspond to the average fluorescence intensityon a given surface crystallographic feature for each electrodeposition condition.t =|FDC − FSW |√σ2DCNDC+σ2SWNSW(A.1)The standard deviation of the measurements: σDC and σSW are the calculated standarddeviation of two or more replicates. The number replicates (NDC and NSW ) are typically 2which are considerably low sample sizes. The calculated t value is compared to the statisticalt value or tabulated t at a degree of freedom calculated using the following formula:DOF =(σ2DC/NDC + σ2SW /NSW )2(σ2DC/NDC)2NDC−1 +(σ2SW /NSW )2NSW−1 (A.2)If the calculated t value is lower than the statistical t value from the table, then FDC andFSW are not significantly different. As a result of the sizeable standard deviations and the lownumber of replicates, the calculated degrees of freedom is lower than the number of replicateseven though the number of replicates exceed two samples. Therefore the statistical t value,obtained for this calculated degree of freedom, which we compare to our calculated t to isgenerally high.The calculated t values compared to their tabulated t values can be seen betweenEdep=0.4Vand ESWdep =0.4V/-0.3V(50 Hz) for both thiol-exchange in Table A.1 and DNA adsorption in Table181Appendix A. T-test Comparison of DNA CoveragesAu(hkl) Calc.tTab. t(95% CI)Tab. t(90% CI)Tab.(80% CI)DOF Fl. Int.Edep(Avg± σv)Fl. Int.ESWdep(Avg± σv)111 3.47 12.706 6.314 3.133 1 2758±3001946±138100 0.95 12.706 6.314 3.133 1 1498±885881±230110 1.96 12.706 6.314 3.133 1 1408 ±951019 ±264311 1.03 4.303 2.920 1.883 2 5643 ±2605347 ±353210 1.19 12.706 6.314 3.133 1 412 ±1961180 ±891Table A.1: Example of calculated T’s for comparison of the Potential-assisted Thiol Exchangeat Edep =0.4V and ESWdep = 0.4V/-0.3V(50 Hz). These conditions had 2 replicates. These resultscorrespond to Figure 6.3A.2. Tabulated t values shown in bold indicate that they are smaller than the calculated t. Thismeans for a given confidence interval, the fluorescence intensity on a given surface crystallo-graphic regions on sames made at Edep and ESWdep are significantly different. From the values inthe bar graphs (Figure6.3 and 6.12), only small changes were observed in DNA SAMs madeat Edep=0.4V and ESWdep =0.4V/-0.3V(50 Hz) which agrees with the results we obtain from thet-tests.182Appendix A. T-test Comparison of DNA CoveragesAu(hkl) Calc.tTab. t(95% CI)Tab. t(90% CI)Tab. t(80% CI)DOF Fl. Int.Edep(Avg± σv)Fl. Int.ESWdep((Avg±σv)111 3.65 4.303 2.920 1.883 2 3776 ±2624648 ±128100 1.26 12.706 6.314 3.133 1 4714 ±1474976 ±254110 1.09 12.706 6.314 3.133 1 4877 ±1284321 ±712311 2.27 4.303 2.920 1.883 2 3111 ±2034096 ±579210 0.36 12.706 6.314 3.133 1 4572 ±1024762 ±731Table A.2: Example of calculated T’s for comparison of the Potential-assisted DNA Adsorptionon bare Au at Edep=0.4V and ESWdep = 0.4V/-0.3V(50 Hz). These conditions had 2 replicates.These results correspond to Figure 6.8183Appendix BDemonstration on the NoisePropagation during Image AnalysisFigure B.1: (Top row) Demonstration of the effect of noise propagation when calculating thefluorescence modulation ((Fmax − Fmin)/Fmax) of a particular region of interest on a DNASAM. (Bottom row) Images calculated after the initial images were despeckled and Gaussianblurred results in reduction of this noise propogation.184Appendix B. Demonstration on the Noise Propagation during Image AnalysisFigure B.2: A comparison of the histogram for the intensities from the calculated fluorescencemodulation ((Fmax − Fmin)/Fmax) image in the previous figure. Without any despeckling andGaussian blurring of the initial images, the histogram of resulting intensities has a wide stan-dard deviation (above). Additionally, some pixels are lost as the calculation of noisy pixelsresults in NaN. As a result of the despeckling and Gaussian blurring of the initial images, thehistogram (bottom) of resulting intensity values has a lower standard deviation. Only valuesbetween 0 and 1 are shown.185Appendix CSEM Images of DNA ClustersThis section shows the SEM images which support the assertion of DNA cluster formationshown in Chapter 7. A high resolution in-situ observation of the existance of DNA clusterswould be ideal, but challenging. The presence of these clusters, and their size and distributioncan be realized using SEM under very specific conditions.207 Briefly, the measurement of DNASAMs on gold needed to be accomplished on atomically smooth surfaces since significanttopological changes would dominate the secondary electron image. This is possible using the111 facets on the gold single crystal electrodes.207Two samples were prepared using either Edep or ESWdep deposition in phosphate sulfate IBcontaining 0.25 μM DNA. The fluorescence modulation was measured before SEM analysisand indicates a larger mobility for the DNA SAM prepared using Edep as compared to theESWdep deposition in agreement with previous results. SEM images of the 111 facet from thesetwo electrode surfaces are shown in Figure C.1. These DNA SAMs were manufactured bythe thesis author and the SEM images were collected by Isaac Martens. The increase inFigure C.1: SEM images of DNA SAMs on Au(111) manufactured at Edep =+0.4 V/SCE; c)ESWdep = +0.4 V to -0.3 V/SCE (50Hz)l186Appendix C. SEM Images of DNA Clusterssecondary electron signal (white) results from the presence of counter ions decorating theDNA strands.207 The SEM resolution is not high enough to see the position of individual DNAadsorbates. This results in images that show a more uniform distribution of the signal asseen in the Edep example (Figure C.1a). On the other hand, DNA clustering resulted in thepresence of more salt located at or around the cluster which was observed in the SEM withthe resolution used here. Figure C.1b shows evidence of the location and approximate size ofthe DNA clusters which formed on the 111 facet when prepared using ESWdep .187


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items