Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Adsorption of a carboxylated silane on gold : characterization and application to PDMS-based electrochemical… Yang, Cheng Wei Tony 2016

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

Item Metadata

Download

Media
24-ubc_2016_september_yang_tony.pdf [ 20.28MB ]
Metadata
JSON: 24-1.0308091.json
JSON-LD: 24-1.0308091-ld.json
RDF/XML (Pretty): 24-1.0308091-rdf.xml
RDF/JSON: 24-1.0308091-rdf.json
Turtle: 24-1.0308091-turtle.txt
N-Triples: 24-1.0308091-rdf-ntriples.txt
Original Record: 24-1.0308091-source.json
Full Text
24-1.0308091-fulltext.txt
Citation
24-1.0308091.ris

Full Text

Adsorption of a Carboxylated Silane onGold: Characterization and Applicationto PDMS-Based Electrochemical CellsbyCheng Wei Tony YangB.Sc., The University of British Columbia, 2008A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Chemical and Biological Engineering)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)August 2016© Cheng Wei Tony Yang 2016AbstractIntegrated sensing and biosensing microfluidic systems often require sealing between polydimethylsilox-ane (PDMS), glass, and gold interfaces. Studying substances that can self-organize onto glass and goldsurfaces may achieve these goals and pave the way for new technological advances. Work presentedin this thesis focuses on characterizing the adsorption of N-[(3-trimethoxysilyl)propyl]ethylene-diaminetriacetic acid (or TMS-EDTA) on Au and applying this knowledge to construct leak-free PDMS-basedelectrochemical cells.First, surface analysis of TMS-EDTA-modified Au surfaces was conducted using various techniques.Water contact angle measurements and X-ray photoelectron spectroscopy confirm that the carboxylatedsilane can chemically modify Au surfaces. Atomic force microscopy studies indicate that a uniform sur-face coverage with monolayer thickness is formed. Infrared spectroscopy studies indicate that there islittle evidence of siloxane cross-linking. Surface plasmon resonance results suggest that the carboxy-lates on TMS-EDTA-modified Au are available for streptavidin immobilization.Second, electrochemistry was used to determine the Gibbs free energies of adsorption of TMS-EDTAon Au under aqueous conditions. Electrochemical differential capacitancemeasurements reveal that thepotential-dependent free energies of adsorption are ∼ −20 to −30 kJ/mol (for potentials between −0.5and 0.2 V) in the complex electrolyte solution used. Furthermore, at highly negative potentials (∼−1.1V), TMS-EDTA adsorbs minimally onto the Au surface.Third, PDMS surfaces were functionalized to present primary amino groups, and glass or gold slideswere functionalized to present carboxyl groups. Strong bonding was achieved by bringing the two sur-faces in contact and reacting at room temperature. Shear tests reveal that the novel carboxyl-aminebonding strategy achieved a comparable bond strength as the conventional methods. Subsequently,TMS-EDTA was applied to construct leak-free PDMS-based electrochemical cells. Pressure leak testswere conducted to provide a more realistic measure of the bond strengths under aqueous conditions.A method to electrochemically remove the adsorbed TMS-EDTA layer off of the Au electrode, whilemaintaining the sealed cell chamber, was also developed.iiAbstractThe characterization studies and fabrication strategy presented have led to the development of leak-free PDMS-based electrochemical devices that are suitable for sensing and biosensing applications.iiiPrefaceParts of this thesis have been published as three peer-reviewed journal articles:• Article #1 (http://pubs.acs.org/doi/abs/10.1021/ac902926x) is published by American ChemicalSociety.1 Drs. David Ng and Joanne Fox conceived of the initial idea. As first author, I designed,performed, and analyzed the research with some students. Jake Abbot and Cameron Lawson(high school students at the time) developed the initial protocol and tested the first Jell-O chip.Adrian Lee (undergraduate student at the time) created new experiments based on their tech-niques. I drafted the initial manuscript, and Eric Ouellet made contributions to the text and figures.Dr. Eric Lagally provided guidance throughout the manuscript submission.– Yang, C. W. T.; Ouellet, E.; Lagally, E. T. Using Inexpensive Jell-O Chips for Hands-OnMicrofluidics Education. Anal. Chem. 2010 82, 5408−5414.• Article #2 (http://pubs.acs.org/doi/abs/10.1021/la1012582) is published by American Chemical So-ciety.2 As co-first authors, Eric Ouellet and I contributed equally to the work. I designed, performed,and analyzed all of the shear tests with technical advice from George Lee (Wood Science Depart-ment, UBC Faculty of Forestry). Tao Lin and Lee Ling Yang assisted with the sample prepara-tion. The X-ray photoelectron spectroscopy (XPS) measurements were conducted in the Inter-facial Analysis & Reactivity Laboratory (IARL) at Advanced Materials and Process EngineeringLaboratory (AMPEL, UBC Vancouver) with the help of Dr. Ken Wong. The water contact anglemeasurements were conducted in the Life Sciences Institute (LSI, UBC Vancouver) with the help ofDr. Johan Janzen. I drafted the initial manuscript, and Eric Ouellet made significant contributionsto the text and figures. Dr. Eric Lagally provided guidance throughout the research and manuscriptsubmission.– Ouellet, E.*; Yang, C. W. T.*; Lin, T.; Yang, L. L.; Lagally, E. T. Novel Carboxyl-Amine Bond-ing Methods for Poly(dimethylsiloxane)-Based Devices. Langmuir 2010, 26, 11609−11614.*These authors contributed equally to the work presented.ivPreface• Article #3 (http://pubs.acs.org/doi/abs/10.1021/acs.jpcc.5b09915) is published by American Chem-ical Society.3 As first author, I designed the research together with Dr. Dan Bizzotto, and performedall of the experimental work and analyses as described. Isaac Martens contributed to the atomicforce microscopy (AFM) experiments and the initial analysis of AFM contact mode imaging data.I drafted the initial manuscript. Isaac Martens and Drs. Elöd Gyenge, Robin Turner, and DanBizzotto made valuable contributions to the manuscript revisions. Dr. Bizzotto provided guidancethroughout the manuscript submission process.– Yang, C. W. T.; Martens, I; Gyenge, E. L.; Turner, R. F. B.; Bizzotto, D. Adsorption of a Car-boxylated Silane on Gold: Characterization for Its Rational Use in Hybrid Glass/Gold Sub-strates. J. Phys. Chem. C 2016, 120, 2675–2683.Please see below for details regarding how these publications, along with additional unpublished exper-imental data, contribute to the construction of this thesis:Chapter 1. The first paragraph of the introductory text in Section 1.1 is reproduced with permissionfrom Yang et al. (2010). The rest of the introductory text in Section 1.1 is reproduced with permissionfrom Ouellet et al. (2010). Portions of the introductory text in Sections 1.1 and 1.2 (and the designof a hypothetical polymer-bonded electrochemical cell as shown in Figure 1.2) are reproduced withpermission from Yang et al. (2016).Chapter 3. Figures 3.2 and 3.3 (and the related text) are reproduced with permission from Ouelletet al. (2010). Figures 3.4, 3.10, and 3.12 (and the related text) are reproduced with permission fromthe main text of Yang et al. (2016). Figures 3.5, 3.6, 3.7, 3.8, 3.9, and 3.11 (and the related text) areexpanded with permission from the Supporting Information of Yang et al. (2016).Chapter 4. Figures 4.1, 4.2, 4.3, 4.4, and 4.5 (and the related text) are expanded with permissionfrom the Supporting Information of Yang et al. (2016). Figures 4.6, 4.7, 4.8, and 4.9 (and the relatedtext) are reproduced with permission from the manuscript.Chapter 5. Figures 5.1, 5.3, 5.4, and 5.5 (and the related text) are reproduced with permission fromOuellet et al. (2010). Table 5.1 and Figures 5.9, 5.10, 5.11, 5.6, 5.12, and 5.13 (modified from the originalfigures created by Josiah To) are reproduced with permission from unpublished experimental results.Pressure leak tests and electrodesorption experiments were developed, conducted, and analyzed incollaboration with Josiah To and Dr. Jannú Casanova-Moreno from Dr. Karen Cheung’s Laboratory(UBC Vancouver), with valuable input from Drs. Dan Bizzotto and Karen Cheung. An alternate versionof the related text and figures is available in Josiah’s M.A.Sc. thesis.4vTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiNomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxivAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvii1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Identifying the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 A Potential Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Scope of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Literature and Background Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.1 Introduction to Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.1.1 Chemical Surface Modification Using SAMs . . . . . . . . . . . . . . . . . . . . . 102.1.1.1 Thiols for Gold Modification . . . . . . . . . . . . . . . . . . . . . . . . . . 102.1.1.2 Silanes for Glass Modification . . . . . . . . . . . . . . . . . . . . . . . . 112.1.2 Polydimethylsiloxane (PDMS) and Biosensors . . . . . . . . . . . . . . . . . . . . 132.1.3 An Example: Surface Plasmon Resonance (SPR) Biosensor . . . . . . . . . . . 152.1.3.1 Carbodiimide Activation Chemistry . . . . . . . . . . . . . . . . . . . . . 162.1.3.2 PDMS-Based SPR Biosensor . . . . . . . . . . . . . . . . . . . . . . . . 16viTable of Contents2.2 Surface Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2.1 Water Contact Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2.2 Atomic Force Microscopy (AFM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.2.2.1 AFM Contact-Mode Imaging . . . . . . . . . . . . . . . . . . . . . . . . . 202.2.2.2 AFM Force-Distance Curve . . . . . . . . . . . . . . . . . . . . . . . . . 212.2.3 Infrared (IR) Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.2.3.1 ATR-FTIR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222.2.3.2 PM-IRRAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232.2.4 X-ray Photoelectron Spectroscopy (XPS) . . . . . . . . . . . . . . . . . . . . . . . 252.2.5 Surface Analysis of Silanes on Gold . . . . . . . . . . . . . . . . . . . . . . . . . . 262.2.5.1 Amino-Silane for Gold Modification . . . . . . . . . . . . . . . . . . . . . 262.2.5.2 Alkylsilanes for Gold Modification . . . . . . . . . . . . . . . . . . . . . . 272.3 Thermodynamics of Adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292.3.1 The Gibbs Adsorption Isotherm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302.3.2 Langmuir Adsorption Isotherm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312.3.3 Standard Gibbs Free Energy of Adsorption ΔG° . . . . . . . . . . . . . . . . . . . 322.3.4 Frumkin Adsorption Isotherm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332.3.5 ΔG◦dsin a Complex Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . 332.4 Electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.4.1 Electrical Double Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.4.1.1 Helmholtz Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.4.1.2 Gouy-Chapman Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.4.1.3 Stern Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.4.1.4 The Role of the Solvent at the Interface . . . . . . . . . . . . . . . . . . 362.4.2 Adsorption on an Electrode Surface . . . . . . . . . . . . . . . . . . . . . . . . . . 372.4.2.1 Electrocapillary Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.4.2.2 The Parallel-Plate Model of Frumkin . . . . . . . . . . . . . . . . . . . . 392.4.2.3 Parabolic Dependence of ΔG◦ds. . . . . . . . . . . . . . . . . . . . . . 402.4.2.4 Pseudocapacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402.4.2.5 Electrodesorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412.5 Integration of Microfluidics & Electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . 422.5.1 Problems with Electrolyte Leaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43viiTable of Contents2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 Surface Analysis of TMS-EDTA Adsorption on Gold . . . . . . . . . . . . . . . . . . . . . . 443.1 Synopsis2,3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443.2 Surface Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443.2.1 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443.2.1.1 General Materials and Gold Surface Preparation . . . . . . . . . . . . . 443.2.1.2 Water Contact Angle and X-ray Photoelectron Spectroscopy (XPS) . . 453.2.1.3 Atomic Force Microscopy (AFM) . . . . . . . . . . . . . . . . . . . . . . . 463.2.1.4 Infrared Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473.2.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483.2.2.1 Water Contact Angle and X-ray Photoelectron Spectroscopy (XPS) . . 483.2.2.2 Atomic Force Microscopy (AFM) . . . . . . . . . . . . . . . . . . . . . . . 483.2.2.3 Infrared Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533.3 Surface Plasmon Resonance (SPR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613.3.1 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613.3.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 623.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624 Thermodynamic Studies of TMS-EDTA Adsorption on Au . . . . . . . . . . . . . . . . . . 654.1 Synopsis3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654.2.1 General Materials and Gold Surface Preparation . . . . . . . . . . . . . . . . . . 654.2.2 Electrochemical Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.2.2.1 Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.2.2.2 Working Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.2.2.3 Differential Capacitance Measurements . . . . . . . . . . . . . . . . . . 664.2.2.4 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.3.1 Open-Circuit Potential (OCP) During Surface Modification . . . . . . . . . . . . . 674.3.2 Initial Measurements of Au Electrochemistry in Working Buffer . . . . . . . . . . 684.3.3 Data Treatment and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694.3.4 Estimation of Surface Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69viiiTable of Contents4.3.5 Determining the Gibbs Free Energies of TMS-EDTA Adsorption onto Au . . . . 734.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 785 Fabrication of PDMS-Based Electrochemical Cells . . . . . . . . . . . . . . . . . . . . . . . 795.1 Synopsis2,4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795.2 Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding Strategy . . . . . . . 805.2.1 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805.2.1.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805.2.1.2 Fabrication of PDMS Slabs (Planar PDMS Surfaces) . . . . . . . . . . 805.2.1.3 Fabrication of Gold-Coated Glass Substrates . . . . . . . . . . . . . . . 805.2.1.4 Cleaning of Glass or Gold-Coated Glass Substrates . . . . . . . . . . . 825.2.1.5 Surface Functionalization of Glass or Gold-Coated Glass Substrate . 825.2.1.6 Irreversible Bonding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 825.2.1.7 Shear Tests of PDMS Slabs Bonded to Glass or Gold-Coated Glass Sub-strates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835.2.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835.3 Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cells . . . 885.3.1 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 915.3.1.1 Fabrication of PDMS Cell Chambers . . . . . . . . . . . . . . . . . . . . 915.3.1.2 Fabrication of Gold-Coated Glass Substrates . . . . . . . . . . . . . . . 925.3.1.3 Fabrication of the 3-Electrode Substrates . . . . . . . . . . . . . . . . . 925.3.1.4 Modified RCA Cleaning Method . . . . . . . . . . . . . . . . . . . . . . . 955.3.1.5 Carboxyl-Amine Bonding Strategy Using TMS-EDTA . . . . . . . . . . 955.3.1.6 Pressure Leak Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 965.3.1.7 Electrodesorption of TMS-EDTA from Gold . . . . . . . . . . . . . . . . 995.3.2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1005.3.2.1 Pressure Leak Tests with Bonded PDMS Cell Chambers . . . . . . . . 1005.3.2.2 Electrodesorption of TMS-EDTA from Gold . . . . . . . . . . . . . . . . 1025.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1056 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1076.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1076.2 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110ixTable of Contents6.3 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115AppendicesA Appendix for Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128B Appendix for Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129C Appendix for Chapter 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140xList of Tables2.1 A list of common functional groups on silicon. . . . . . . . . . . . . . . . . . . . . . . . . . 125.1 Results from pressure leak tests for air-plasma and carboxyl-amine bonding strategies,with the internal pressure (gauge) of the cell chamber shown. Three replicates wereperformed for each bonding strategy. Standard deviations are reported. . . . . . . . . . 100B.1 The algorithm used to collect the electrochemical differential capacitance measurements(using NOVA 1.8 software) at one potential of interest (POI). . . . . . . . . . . . . . . . . 130C.1 Results from shear tests of glass−PDMS bonding strategies. . . . . . . . . . . . . . . . . 140C.2 Results from shear tests of gold−PDMS bonding strategies. . . . . . . . . . . . . . . . . . 140C.3 Results from manual peel tests of gold−PDMS bonding strategies. . . . . . . . . . . . . 141xiList of Figures1.1 (A) Photograph of a PDMS microfluidic chip bonded to a gold-patterned glass slide forSPR imaging applications (inset: a close-up of the marked region of the gold microarray;scale bar: 650 μm). (B) Cross-sectional schematic view of an individual PDMS flow cell,such as the one indicated by the red region selected in (A). It is challenging to form astrong bond between PDMS and glass while maintaining a functional SAM on Au. . . . 31.2 (A) Hypothetical PDMS-based electrochemical cell (with two Au electrodes sputtered onglass). Working electrode = WE, and secondary electrode = SE. (B) Top view of thedevice with the black dotted lines showing the cross sections observed in (C) and (D). (C)Cross-sectional view of the hybrid glass/gold substrate across the TMS-EDTA-modifiedWE inside the PDMS microfluidic cell chamber. (D) Cross-sectional view of the device,illustrating how TMS-EDTA can be used to chemically modify both glass and gold surfacesfor carboxyl−amine bonding with 3-APTMS-modified PDMS. . . . . . . . . . . . . . . . . 52.1 Schematic of a typical biosensor operation (left) and regeneration (right). . . . . . . . . 82.2 Schematic of the three popular immobilization strategies for gold surfaces: physical, co-valent, and bioaffinity immobilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.3 Schematic diagram of an ideal SAM formed on a gold surface. The structure and char-acteristics of the SAM are highlighted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.4 The simplest scheme of forming a SAM on Au: (i) initial physisorption, (ii) chemisorptionof the molecules, (iii) nucleation of the standing up phase, and (iv) formation of highlyordered layer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.5 The reaction mechanism of (A) hydrolysis, (B) condensation and (C, D) bonding of silanesto a hydroxyl-containing surface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.6 The chemical structure of PDMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13xiiList of Figures2.7 Scheme for producing PDMS chips using soft lithography. (A) A mold is made with thedesired features. (B) Liquid pre-polymer material is poured onto the mold. (C) Mold withliquid PDMS is cured at low temperatures. (D) Solidified and flexible PDMS is peeled offand (E) placed on a rigid substrate for experiments. . . . . . . . . . . . . . . . . . . . . . . 142.8 A typical setup of an SPR biosensor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.9 The chemical reaction scheme for the covalent immobilization of a probe onto the sur-face via carboxyl-amine coupling chemistry. (1) The carboxyl group is activated withEDC/NHS. (2) Covalent attachment of a probe by its primary amino group. (3) The unre-acted active site can be deactivated with ethanolamine. . . . . . . . . . . . . . . . . . . . 162.10 The design of a PDMS-based microfluidic device for SPRi applications. . . . . . . . . . 172.11 Illustration of contact angles formed by sessile liquid drop on a solid surface. . . . . . . 182.12 Schematic diagram showing the setup of a typical AFM experiment. . . . . . . . . . . . . 202.13 A schematic of force-distance curve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212.14 Illustration of the phase change upon reflection at a metal surface for the components ofthe incident radiation polarized perpendicular (s) and parallel (p) to the plane of reflection. 232.15 Dependence of the absorption factor at the wavelength of maximum absorption on θ fora 1 nm layer of acetone on metal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242.16 Reflection-absorption Fourier transform infrared spectroscopy (RAIR) spectra of 3-APTESadsorbed on gold from methanol solution, as a function of hydrolysis and condensationreactions. (1) Freshly deposited 3-APTES at room temperature, (2) plus 4 min at 75 °C,(3) plus 6 min, (4) plus 6 min, (5) plus 20 min, and (6) after heating for 12 h at 75 °C. . 262.17 RAIR spectra for monolayers of (A) octadecylsilane, (B) octylsilane, and (C) hexylsilaneon Au. A solution IR spectrum of octylsilane (D) is included for comparison. . . . . . . . 272.18 XPS spectra for (A) C 1s core-levels, and (B) Si 2p and Au 4f core-levels of (1) octade-cylsilane, (2) octylsilane, and (3) hexylsilane chemisorbed on (4) freshly evaporated Au. 282.19 The proposed reaction mechanism of a monolayer of silane on Au. . . . . . . . . . . . . 282.20 (A) O 1s and (B) C 1s core levels of an octylsilane monolayer on gold exposed to succes-sive dose of O3/O2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292.21 A schematic representation of the electrical double layer of a negatively charged inter-face. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37xiiiList of Figures3.1 Chemical structures of the molecules used in this chapter: TMS-EDTA, 11-MUA, EDTA,and 3-APTMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453.2 Water contact angle measurements on (A) bare gold, (B) 11-MUA-modified gold, and (C)TMS-EDTA-modified gold. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493.3 XPS high resolution spectra of the C 1s signal for (A) 11-MUA and (B) TMS-EDTA on gold(with spectrum for bare gold surface). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503.4 Topography (A) and lateral friction (B) maps from AFM contact mode imaging are shown(scanned using the same tip for all images). Bare Au substrate (top row); TMS-EDTA-modified Au (middle row); and the bare Au substrate after scanning the TMS-EDTA layer(bottom row). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513.5 Topography (A), and lateral friction (B) distributions calculated from AFM contact modeimages (scanned using the same tip for all images): cleaned gold substrate before samplescan (red), TMS-EDTA modified gold substrate scan (black), and cleaned gold substrateafter sample scan (blue). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523.6 (A) An example of the 64 approach-only AFM force-distance curves for TMS-EDTA-modifiedAu. (B) First derivatives of the force-distance curves. . . . . . . . . . . . . . . . . . . . . . 543.7 (A) The maximum of the derivatives was defined as the surface of Au (i.e., distance = 0nm). The red curve shows the averaged values for the 64 samples. (B) A zoomed-inview of the averaged and normalized values, with the blue line serving as the referenceline. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553.8 The zoomed-in view of silicon nitride tip interaction with TMS-EDTA-coated gold (red),11-MUA-coated gold (blue), and clean gold substrate (black). . . . . . . . . . . . . . . . . 563.9 An example of background correction method used for PM-IRRAS data analysis. (A) Theraw signal of TMS-EDTA-modified Au (solid line) is shown with background curve deter-mined from spline fitting (dotted line). (B) The background-subtracted signal is shown.Subsequently, this signal was corrected for sample gain(s) to obtain Absorbance spec-trum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573.10 The PM-IRRAS spectra for 3-APTMS on Au (black), 11-MUA on Au (red), and TMS-EDTA on Au (blue). Alkyl stretching region (3200–2700 cm−1) and the fingerprint region(1800–800 cm−1) are shown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58xivList of Figures3.11 Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) trans-mission spectra of bulk TMS-EDTA (blue), EDTA (green), and 3-APTMS (red) mixed withKBr powder. (A) Alkyl stretching region (3200–2700 cm−1) and (B) the fingerprint region(1800–800 cm−1) are shown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593.12 (A) SPR sensorgram showing the immobilization of streptavidin on a bare Au chip, fol-lowed by five injections of the 50 mM NaOH regeneration buffer. (B) Streptavidin im-mobilization on TMS-EDTA-modified Au chip using NHS/EDC chemistry, followed by fiveinjections of the same regeneration buffer. . . . . . . . . . . . . . . . . . . . . . . . . . . . 634.1 The open-circuit potential (OCP) of bare Au bead electrode immersed in 10% TMS-EDTA(v/v) solution with (dotted curve) and without (solid curve) oxygen were determined. . . 684.2 (A) Raw capacitance data for DiffCap Buffer measurements from four independent ex-periments (from four different days). (B) Averaged raw DiffCap Buffer data and its corre-sponding standard deviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 704.3 The stability of freshly cleaned bare Au bead electrode (circle) and TMS-EDTA-modifiedAu bead electrode (square) in DiffCap buffer at open-circuit potential (OCP). . . . . . . 714.4 Raw capacitance data for three independent measurements (from three different days) of20 μM TMS-EDTA in DiffCap buffer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 714.5 Averaged and normalized electrochemical differential capacitance data at equilibrium(360 s with stirring) for individual DiffCap buffer components: (1) 100 mM phosphate (cir-cle), (2) 100 mM phosphate and 65 mM KOH (square), and (3) DiffCap Buffer (triangle)– 100 mM phosphate, 65 mM KOH, and 500 mM KCl. . . . . . . . . . . . . . . . . . . . . 724.6 Capacitance–potential curves of increasing TMS-EDTA concentrations in DiffCap Buffer(select curves shown). For each concentration, capacitance was measured after eachpotential was held for 6 min with stirring. Series resistor–capacitor equivalent circuit wasassumed for calculations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 734.7 (A) The fractional surface coverage curves calculated from capacitance values using Cθ=Cθ=0(1− θ)+Cθ=1θ. (B) A limited potential window is shown. . . . . . . . . . . . . . . 744.8 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of (A) −0.5 V, (B)−0.25 V, and (C) 0.1 V. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75xvList of Figures4.9 Potential-dependent (A) free energies of adsorption and (B) lateral interaction parameterα, which are determined from Frumkin isotherm fitting. Positive α indicates repulsiveinteraction. Continuous interpolated values (solid line) are shown. Errors are estimatedby robust fitting routine in MATLAB and represent 95% confidence interval. . . . . . . . 775.1 Chemical structures of the molecules used in shear tests: 3-APTMS, 3-MPTMS, BTMSE,TMS-EDTA, and 11-MUA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815.2 Scheme for producing PDMS slabs using soft lithography. (A) A flat aluminum weighingpan is used as the mold. (B) Liquid PDMS is poured onto the mold. (C) Mold with liquidPDMS is baked in an oven. (D) Solidified PDMS is peeled off and (E) cut into slabs ofappropriate sizes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815.3 Water contact angle measurements on (A) bare PDMS, (B) PDMS exposed to UV-ozone(UVO) for 5 min, and (C) 3-APTMS-modified PDMS. . . . . . . . . . . . . . . . . . . . . . 845.4 (A) Modification of PDMS with 3-APTMS to form primary amines. (B) Modification of goldwith 11-MUA to form carboxylic acids. (C) Carbodiimide activation of the carboxylic acidgroups, followed by irreversible bonding of PDMS to gold by physical contact at roomtemperature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 865.5 (A) Bond failure curves of TMS-EDTA-modified glass and 3-APTMS-modified PDMS slabs.The peak of each curve represents the failure point of the PDMS−substrate bond. (B)Bond strengths of four different PDMS−glass bonding strategies: UV-ozone treatment ofboth glass and PDMS surfaces (UVO); oxygen-plasma treatment of both glass and PDMSsurfaces (O2Plasma); TMS-EDTA-modified glass with 3-APTMS-modified PDMS (TMS-EDTA & 3-APTMS); and BTMSE-modified glass with UV-ozone-treated PDMS (BTMSE& UVO). (C) Bond strengths of three different PDMS−gold bonding strategies: 3-MPTMS-modified gold with UV-ozone-treated PDMS (3-MPTMS & UVO); 11-MUA-modified goldwith 3-APTMS-modified PDMS (11-MUA& 3-APTMS); and TMS-EDTA-modified gold with3-APTMS-modified PDMS (TMS-EDTA & 3-APTMS). All samples were analyzed in tripli-cate. Error bars represent standard deviation. . . . . . . . . . . . . . . . . . . . . . . . . . 875.6 Schematic of the fabricated 3-electrode device. (a) Top view of the PDMS-bonded device.(b) Cross-sectional view of the device through the center (i.e., down the length) of thedevice. (c) Enlarged view of the proposed carboxyl-amine chemistry between PDMS andthe gold/glass substrate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90xviList of Figures5.7 Step 1 of 2: Scheme for producing PDMS cell chambers from 3D-printed molds. (A) A3D-printed mold is made with desired features. (B) Liquid PDMS is poured onto the mold.(C) Mold with liquid PDMS is cured. (D) Solidified PDMS is peeled off. The rough surfaceof the 3D-printed mold resulted in a rough surface on the PDMS cell chamber produced. 925.8 Step 2 of 2: Scheme for producing the final PDMS cell chambers using glass/PDMSmolds. (A) PDMS cell chamber made with the 3D-printed mold is used as the initial mold.(B) Liquid PDMS is poured into the hollow chamber. (C) Mold with liquid PDMS is curedand a small disk is produced. (D) Small disk is bonded to a clean glass slide by air-plasma.(E) Liquid PDMS is poured onto the new glass/PDMS mold. (F) Mold with liquid PDMSis baked. (G) Solidified PDMS cell chamber is peeled off and cut into appropriate sizes. 935.9 Molds used for creating PDMS cell chambers with smooth surfaces. (A) Designs of the3D-printed molds for creating the initial PDMS cell chambers (Step 1), from which thesmall PDMS disks were created. (B) Small PDMS disks were then bonded to clean glassslides by air-plasma in order to create the final mold design. The final PDMS cell chamberswere fabricated using these PDMS/glass molds (Step 2). . . . . . . . . . . . . . . . . . . 945.10 Design of the 3-electrode substrate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955.11 Setup used to evaluate the strength of the bond (via leak pressure tests) between PDMSand solid substrates (glass, gold, and mixed Au/glass substrates). . . . . . . . . . . . . . 985.12 (A) PDMS cell chamber bonded to the mixed Au/glass substrate using the air-plasmabonding strategy right after loading (i.e., at atmospheric pressure). (B) Initial leak pressureof the plasma-bonded device (i.e., 8 kPa). (C) Complete failure of the plasma-bondeddevice (i.e., 29 kPa). PDMS bonded to the 3-electrode substrate using the carboxyl-aminebonding strategy at (D) the atmospheric pressure after loading, (E) the leak pressure ofplasma-bonded device (i.e., 8 kPa), and (F) the pressure of an initial solution leak (i.e.,49 kPa). Diameter of the cell chamber was 7.6 mm. . . . . . . . . . . . . . . . . . . . . . 1015.13 CoHex cyclic voltammograms of TMS-EDTA covered Au. (A) Three different methods toclean a gold bead are shown: pulsing to negative potentials (-1.4 V) in pH 13 solution,CV in 1 M sulfuric acid, and flaming. (B) Two different methods to clean the Au within abonded PDMS device are shown: pulsing to negatives potentials in basic solution andperforming CV in acidic solution. Potential scan rate = 20 mV/s. . . . . . . . . . . . . . . 104A.1 Titration curve for adding 1M HCl to 10% TMS-EDTA (v/v) solution. . . . . . . . . . . . . 128xviiList of FiguresB.1 Full range CV of the DiffCap buffer was obtained. . . . . . . . . . . . . . . . . . . . . . . . 130B.2 Double layer CV of the DiffCap buffer was obtained before and after the addition of bulkTMS− EDTA concentrations for all experiments. Two TMS− EDTA concentrations (20μM and 20 mM) are shown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131B.3 Raw differential capacitance data of DiffCap buffer. . . . . . . . . . . . . . . . . . . . . . . 131B.4 Raw differential capacitance data of 2 μM TMS-EDTA. . . . . . . . . . . . . . . . . . . . . 132B.5 Raw differential capacitance data of 20 μM TMS-EDTA. . . . . . . . . . . . . . . . . . . . 132B.6 Raw differential capacitance data of 200 μM TMS-EDTA. . . . . . . . . . . . . . . . . . . 133B.7 Raw differential capacitance data of 2 mM TMS-EDTA. . . . . . . . . . . . . . . . . . . . . 133B.8 Raw differential capacitance data of 20 mM TMS-EDTA. . . . . . . . . . . . . . . . . . . . 134B.9 Raw differential capacitance data of 200 mM TMS-EDTA. . . . . . . . . . . . . . . . . . . 134B.10 Capacitance curves of 50 mM Perchlorate Buffer (solid line) and DiffCap Buffer (dottedline) at pH 12 are shown. These curves were used to estimate the area of the Au beadin DiffCap Buffer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135B.11 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −1.1 V. . . . 136B.12 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −1.0 V. . . . 136B.13 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −0.9 V. . . . 137B.14 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −0.8 V. . . . 137B.15 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −0.7 V. . . . 138B.16 Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dottedline) and Frumkin (solid line) isotherms, at applied electrode potentials of −0.6 V. . . . 138B.17 The potential-dependent R2 values from the Langmuir non-linear least squares fitting. 139B.18 The potential-dependent R2 values from the Frumkin non-linear least squares fitting. . 139C.1 (A) A schematic of a typical sample used for shear tests: a PDMS slab bonded to a glasssubstrate is shown. (B) Top view of the PDMS-bonded glass substrate. (C) A side-viewdiagram of the shear test procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142xviiiList of FiguresC.2 An example of pressure leak test experiment for carboxyl-amine bonded PDMS-based3-electrode device. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143xixNomenclature3D three-dimensionalAFM atomic force microscopyATR attenuated total reflectanceATR-FTIR attenuated total reflectance Fourier transform infrared spectroscopyCE counter electrodeCV cyclic voltammetryDiffCap DiffCap buffer used for differential capacitance experimentsIHP inner Helmholtz planeIR infraredIRRAS infrared reflection-absorption spectroscopyMR magnetic resonanceOHP outer Helmholtz planePET positron emission tomographyPM-IRRAS polarization modulation-infrared reflection-absorption spectroscopyPOI potential of interestpsi pound per square inch or pound-force per square inchPZC point of zero chargeRAIR reflection-absorption Fourier transform infrared spectroscopyRE reference electrodexxNomenclatureSAMs self-assembled monolayersSCE saturated calomel electrodeSE secondary electrodeSPR surface plasmon resonanceSPRi surface plasmon resonance imagingSSCE silver/silver chloride electrodeUHV ultra-high vacuumUV ultravioletWE working electrodeXPS X-ray photoelectron spectroscopy11-MUA 11-mercaptoundecanoic acid3-APTES (3-aminopropyl)triethoxysilane3-APTMS (3-aminopropyl)-trimethoxysilane3-MPTMS (3-Mercaptopropyl)trimethoxysilaneBTMSE 1,2-bis(trimethoxysilyl)ethaneCoHex hexaamminecobalt(III)EDC N-(3-(dimethylamino)propyl)-N´-ethylcarbodiimide hydrochlorideEDTA ethylenediaminetetraacetic acidNHS N-hydroxysuccinimidePDMS polydimethylsiloxaneTMS-EDTA N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acidα Frumkin lateral interaction parameter (unit:kJmo) surface excess of a particular species (unit:mom2)xxiNomenclatureγ surface or interfacial tension (unit:NmorJm2) surface excess of species γLG surface tension of the liquid-gas interfacem surface excess at maximum surface coverageγSG surface tension of the solid-gas interfaceγSL surface tension of the solid-liquid interfaceμ chemical potential of species ν frequency (unit: s−1)σ stored charge density (unit:Cm2)σM excess surface charge density of the metalθ the fraction of surface occupied by the adsorbing moleculeθC contact angleϵ dielectric constant of the medium (or relative permittivity)ϵ0 permittivity of free space (unit:CV ·m)ΔG° standard Gibbs free energy (unit:kJmo)ΔG◦dsstandard Gibbs free energy of adsorptionEb binding energy (unit: eV)Ek kinetic energy (unit: eV)Er recoil energy (unit: eV)A area of the dividing surfaceφsp spectrometer work function (unit: eV)A free molecules (or adsorbates) that can adsorb onto the surfaceAd adsorbed molecules (or adsorbates) on the surfacexxiiNomenclatureC bulk concentration of a particular species (unit:moLor M)Cθ=0 capacitance of the electrode without the molecule of interestCθ=1 capacitance of the electrode completely covered by the molecule of interestCd capacitance of the double layerCD capacitance of the diffuse layerCd differential capacitance (unit:CV ·m2orFm2)CH capacitance of the Helmholtz layerd spacing between the two plates (unit: m)E rational electrical potential with a reference at the point of zero chargeEP peak potentialh Planck constant (6.626×10−34 J ·s)K thermodynamic equilibrium constantk rate constant for adsorptionkd rate constant for desorptionKF Frumkin equilibrium constant (unit:Lmoor M−1)KL Langmuir equilibrium constant (unit:Lmoor M−1)Nƒ ree number of free sitesn the number of moles in excess of species Ns total number of sitesR gas constant (8.314JK ·mo)r rate of adsorptionrd rate of desorptionRp reflectivity due to p-polarized radiationxxiiiNomenclatureRs reflectivity due to s-polarized radiationS free vacant sites on the surfaceT temperature (unit: K)V voltage drop (unit: V)xxivAcknowledgmentsMy academic journey began when I was hired as an Undergraduate Research Assistant by Dr. WunChey Sin (of Dr. Christian Naus Lab) at Life Sciences Centre. I was able to conduct scientific researchunder her guidance and mentorship. Dr. Sin encouraged me to explore some research opportunitiesin the United States. In the summer of 2007, I secured a summer internship position at WashingtonUniversity in St Louis with the Medical Physics Division, under the supervision and mentorship of Dr.Dan Low (now at UCLA). His advice for me was to pursue an Engineering degree to enrich my theoreticalstudies in Physics and Life Sciences. Dr. Sin and Dr. Low: I treasure the life experience you have sharedwith me. Thank you for being passionate about helping young students.In 2008, I was hired by Dr. Eric T. Lagally as the Lab Manager, and officially started my M.A.Sc.degree in 2009 with the Department of Chemical and Biological Engineering (CHBE) at UBC. Dr. La-gally’s "open door" policy, scientific curiosity, encouragement to be innovative and learn from mistakes,and passion for Science and Engineering outreach have forever influenced my academic and personallife. Shortly after, I transferred to the PhD program. Dr. Lagally: I appreciate the opportunities you haveopened up for me.After the decommissioning of Lagally Lab in 2011, Dr. Jim Kronstad - Head of Michael Smith Labo-ratories (MSL) - and Dr. Peter Englezos - Head of CHBE - worked together to revive my PhD program.After consultation with Drs. Eric Lagally, Bhushan Gopaluni, Előd Gyenge, Robin Turner, Dan Bizzotto,and Jamie Piret, my PhD Supervisory Committee was formed:• Előd Gyenge, Co-Supervisor (CHBE): Electrochemistry• Robin Turner, Co-Supervisor (MSL, CHEM, ECE): Surface Plasmon Resonance & Electrochem-istry• Dan Bizzotto, Committee Member (CHEM): Electrochemistry & Physical Chemistry• Jamie Piret, Committee Member (MSL, CHBE): Microfluidics & BioreactorsDr. Piret: Thank you for the technical advice. Dr. Gyenge: Thank you for the invaluable scientific andxxvAcknowledgmentsengineering inputs. I also appreciate your effort in securing some funding fromCHBE for my PhD project,and providing initial funding for my Science and Engineering outreach activities. Dr. Turner: Thank youfor helping me to see the Big Picture during the most difficult of times. Your direct, critical, objective, andtimely advice has helped me tremendously along this tortuous journey. With your continuous support,I have been able to focus on the Science and Engineering aspects of my PhD degree. Dr. Bizzotto:Thank you for everything that you have done for me. This thesis is possible because of your unwaveringpatience, guidance and support. Thank you for standing by my side to run many of the experiments.Thank you for the countless hours of discussions, edits, and submissions. Thank you for providing manyof the resources to help me succeed academically.Furthermore, my academic journey has been thoroughly enjoyable with the company of the followingfriends and colleagues:• Dr. Eric Lagally Lab members: Eric Ouellet, Tony Lam, Jie Yu, Bernard Coquinco, Roza Bidshahri,Louise Lund, Lee Ling Yang, and Antonio Villegas• Dr. Dan Bizzotto Lab members: Jannú Casanova-Moreno, Amanda Musgrove, Landis Yu, IsaacMartens, Santa Maria Gorbunova, Caroline Pao, Kaylyn Leung, Elizabeth Fisher, and Jeff Mirza• Drs. Michael Blades and Robin Turner Lab members: Yan Tan, Chad Atkins, Georg Schulze,Kevin Buckley, and Stanislav Konorov• Dr. Karen Cheung Lab members: Josiah To, Samantha Grist, and Yan Li• PastCHBEGraduate Students Club Executives: HoomanRezaei, RahmanGholami, Siduo Zhang,Fahimeh Yazdanpanah, Alireza Bagherzadeh, Ehsan Behzadfar, and Md. Hafizur Rahman• Past Thunderbird’s Residence Life Team members: Fiona Hess, Jayden Beaudoin, Karen Ratch-ford, Lindsey Curtis, Margarita Iturriaga Bustamante, Poureya Bazargani, Sean Kim, Simi Toma,Yiwei Liu, Cristel Moubarak, KenMcClure, Jenny Huang, Yanlong Guo, Varune Rohan Ramnarine,and Emily Gao• ATP Band members: Ales Horak, Tina Chou, Noriko Tanaka, and Hardy Hall• MSL, AMPEL, and CHBE Faculty, Researchers, Staff, and StudentsLast but not least, I would like to sincerely thank the unconditional love and support from my family:Harry, Sherry, Jack, Sharon, and Margaret. Thank you for everything! To my family around the world: Ilook forward to reuniting with all of you soon.xxviDedicationTo my Family...xxviiChapter 1Introduction1.1 Identifying the ProblemThe interdisciplinary field of microfluidics develops miniaturized technologies for manipulating the flowand reaction of small amounts of fluids.5–7 Microfluidics has the potential to revolutionize modern biologyand medicine because it offers the advantages of working with smaller reagent volumes and shorterreaction times, which greatly reduce the cost required for an analysis.8,9 Current efforts are being madeto integrate an entire laboratory of analytical instrumentation onto a single chip to produce “lab on achip” systems.10,11 Microfluidics has been applied to solve problems in diverse areas in both basic andapplied sciences,12–16 and highly parallelized microfluidic systems are also being actively explored toimprove existing technologies.17,18 Currently, highly integrated microfluidic chips with a thousand ormore detection chambers can be produced and implemented easily.19–21Many types of materials have been introduced for microfluidic chip fabrication.22 Polydimethylsilox-ane (PDMS) is quite attractive for the development of microfluidic applications not only due to its bio-compatibility,23 but also because it is transparent, nontoxic, permeable to gases, thermally stable, andeasy to handle, and it can be used to form submicrometer structures.24–26 The recent advancementsin fast prototyping and mass production technologies, combined with the inexpensive nature of thismaterial, have made this polymer quite attractive in a variety of fields.27–32 For longer than a decade,several PDMS-based medical devices have been under development and used in blood pumps, cardiacpacemaker leads, catheters, drainage tubing, implants, and contact lenses.33Despite its many advantages, however, a major area of concern with the use of PDMS is the bondingof such a polymer to other materials, as this step is crucial for the assembly of microfluidic devices.34 Forinstance, gold surfaces are frequently integrated into sensors and biosensors made of silicon and glassfor the quantification of a variety of substances (see Figures 1.1A and 1.2A).35 Therefore, it is importantto form a good seal simultaneously between PDMS, glass, and gold interfaces for these integratedmicrofluidic systems.11.1. Identifying the ProblemA number of PDMS bonding strategies have been reported. The simplest method involves forming areversible bond using clamps to maintain the seal.36 However, the use of clamps can be cumbersomewhen additional equipment or apparatus is required. Also, the force exerted by the clamp(s) may deformthe PDMS and add internal stress to the device. Oxygen plasma treatment of PDMS surfaces has beenwidely used, whereby silanol groups are created on the PDMS surface and can form covalent siloxanebonds with some substrates (e.g., glass, quartz, or PDMS).37 However, other substrates (e.g., gold) willnot bond by this method. For bonding PDMS to metals (such as gold and silver), mercaptosilanes havebeen widely used in the past38–41 and very few alternatives exist. As a result, an additional passivationlayer (e.g., silicon dioxide or SU-8 photoresist) is often coated on top of the gold substrate, in order touse the plasma bonding strategy.42,43 Other methods that have been used make use of an adhesivelayer on the substrate in order to bond the two surfaces together.44,45 Though this strategy may beeffective, the sensor surface may become inaccessible due to the adhesive layer. Also, the use of eithera passivation layer or an adhesive layer involves additional cleanroom time and requires specializedequipment. Developing a simple and an economical fabrication method, which leads to sensitive, stable,and reliable integrated systems still remains a challenge.46In the case of biosensors, a sensor surface (usually gold) is modified with biomolecular recogni-tion elements (such as proteins, nucleic acids, or cells) to detect a specific target of interest. Forexample, microfluidic surface plasmon resonance (SPR) imaging biosensors often require the use ofglass substrates partially patterned with a microarray of gold spots (see Figure 1.1A).18 The use of self-assembledmonolayers (SAMs) on the sensor surface is common for the immobilization of recognition el-ements.47–49 Commonly, carboxylic acid-terminated alkanethiols (COOH−R−SH) are used to covalentlyattach biomolecules at sites of primary amines using the well-known N-hydroxysuccinimide (NHS) andN-(3-(dimethylamino)propyl)-N´-ethylcarbodiimide hydrochloride (EDC) coupling chemistry.50,51 How-ever, due to the destructive nature of the existing bonding techniques (such as oxygen plasma or UV-ozone), these SAMs and biological recognition elements can be easily degraded or removed from theAu surface. In addition, most alkanethiol molecules are most soluble in organic solvents. Depending onthe incubation time of the organic solvent and the design of the microfluidic chip, the PDMSmaterial mayswell,52–54 causing delamination from the surface and deformation of any channel structures moldedinto the PDMS. Therefore, it is challenging to form a strong bond44 between PDMS and glass while main-taining a compatible surface on the Au (inside the PDMS flow cell) for the covalent immobilization55 ofbiomolecules (see Figure 1.1B).Microfluidic electrochemical devices often utilize glass substrates decorated with sputtered Au elec-21.1. Identifying the ProblemFigure 1.1: (A) Photograph of a PDMS microfluidic chip bonded to a gold-patterned glass slide for SPRimaging applications (inset: a close-up of the marked region of the gold microarray; scale bar: 650μm).18 Reproduced with permission from RSC Publishing. (B) Cross-sectional schematic view of anindividual PDMS flow cell, such as the one indicated by the red region selected in (A). It is challengingto form a strong bond between PDMS and glass while maintaining a functional SAM on Au.31.2. A Potential Solutiontrodes.56 For example, Figure 1.2A shows a hypothetical PDMS-based electrochemical cell (with twogold electrodes sputtered on glass). Figure 1.2B shows the top view of the device with the working elec-trode (WE) and the secondary electrode (SE); such a configuration requires a seal at both PDMS−glassand PDMS−gold interfaces. During analysis, an aqueous solution of analyte(s) is injected into thePDMS microfluidic cell chamber. However, without modifying the gold electrodes, leaks can occur at thePDMS–Au interface, which can compromise the device performance.57 Therefore, studying substancesthat can self-organize onto glass and gold surfaces is required to address some of these issues and topave the way for new technological advances.581.2 A Potential SolutionIn general, it is well known that SAMs with thiols can modify Au or other metal surfaces,59 and thatsilane-based SAMs can modify glass, PDMS or metal oxide surfaces.60 For example, a carboxylic acid-terminated alkanethiol (11-mercaptoundecanoic acid, or 11-MUA) has been widely studied due to its abil-ity to attach strongly to gold and to covalently capture primary amine-terminated biomolecules.61 Sim-ilarly, a carboxylated silane (N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acid or TMS-EDTA)has been widely used to modify oxide-, silanol-, and hydroxyl-containing surfaces and nanoparticles forheavy metal ion chelation,62 PET and MR imaging,63 rare-earth ion adsorption and separation,64,65 andbacteriophage immobilization.66Interestingly, there is a growing literature suggesting that some silanes can also modify Au sur-faces,67–69 resulting in an expansion of the application space for silanes. However, the modificationof Au substrates using TMS-EDTA (i.e., a carboxylated silane) has not been reported in the literature.If TMS-EDTA can modify both glass and Au, then it may offer distinct advantages over 11-MUA, whichcan only modify Au but not glass. More specifically, TMS-EDTA can potentially be used in a one-step,solution-based modification strategy for glass substrates that have been partially patterned with goldelectrodes (to assist in both biosensing and sensor fabrication). To illustrate this concept, Figure 1.2Cshows the cross-sectional view of the mixed Au/glass substrate across the WE inside the PDMS cham-ber. Functional carboxylates are potentially available on the Au electrode (via TMS-EDTA modification).This surfacemay then be used to immobilize biomolecules for biosensing applications. Moreover, Figure1.2D shows the cross-sectional view of the device across all Au electrodes. After chemically modifyingthe surface of PDMS with (3-aminopropyl)-trimethoxysilane (3-APTMS), TMS-EDTA can potentially beused to chemically modify both glass and gold surfaces for carboxyl–amine bonding with primary amine-41.2. A Potential SolutionFigure 1.2: (A) Hypothetical PDMS-based electrochemical cell (with two Au electrodes sputtered onglass). Working electrode =WE, and secondary electrode = SE. (B) Top view of the device with the blackdotted lines showing the cross sections observed in (C) and (D). (C) Cross-sectional view of the hybridglass/gold substrate across the TMS-EDTA-modified WE inside the PDMSmicrofluidic cell chamber. (D)Cross-sectional view of the device, illustrating how TMS-EDTA can be used to chemically modify bothglass and gold surfaces for carboxyl−amine bonding with 3-APTMS-modified PDMS.3 Reproduced withpermission from ACS Publications.51.3. Scope of the Thesisterminated PDMS. Specifically, strong bonds between PDMS–Au and PDMS–glass could seal the cellchamber and prevent leaks.1.3 Scope of the ThesisThe surface modification of Au by TMS-EDTA has not been previously characterized in any rigorousmanner. It has therefore not been possible to exploit the utility of TMS-EDTA to achieve an optimal bondwhile maintaining a chemically modified or electrochemically compatible Au surface. Therefore, thisthesis aims to first address existing knowledge gaps, and then apply the resulting knowledge to facilitatethe design of a leak-free PDMS-based electrochemical cell. Thesis objectives include:• Characterize the adsorption of TMS-EDTA on Au using different surface analysis techniques (Chap-ter 3)• Demonstrate the feasibility of using TMS-EDTA-modified Au to capture biomolecules containingprimary amino groups (Chapter 3)• Quantify the Gibbs free energies of TMS-EDTA adsorption onto Au using electrochemical methods(Chapter 4)• Assess the proposed carboxyl–amine bonding strategy (i.e., 3-APTMS-modified PDMS in contactwith TMS-EDTA-modified glass or Au substrate) using shear tests (Chapter 5)• Apply the knowledge obtained about TMS-EDTA adsorption on Au to develop leak-free PDMS-based electrochemical cells (Chapter 5)6Chapter 2Literature and Background ReviewThis thesis project is multidisciplinary and involves numerous areas of research, including biosensors,surface science, microfluidics and electrochemistry. The objective of this chapter is to provide the readerwith an introduction to the topics, theories, and techniques encountered in this work.2.1 Introduction to BiosensorsBiosensor technology encompasses an expanding field used in diverse applications including health-care diagnostics, forensic analysis, metabolic engineering, environmental monitoring, and food qualityevaluation.70–75 The term “biosensor” indicates a sensing device that integrates three key elements (seeFigure 2.1):76 1) a biomolecular recognition element (i.e., a probe) that is selective towards a particularanalyte molecule (i.e., a target), 2) a selective interaction event between the target and the probe, and 3)a transduction mechanism (i.e., physicochemical change such as current flow, heat transfer, or changein refractive index or mass) to convey the occurrence of this recognition event.77Optical (e.g., colorimetric, fluorescent, and plasmonic) and electrochemical (e.g., amperometric,potentiometric, and impedimetric) transduction mechanisms are often used.78,79 These sensing ap-proaches have been applied in biosensors to analyze a wide range of biomolecular interactions.80–83Typically, the biorecognition element (e.g., protein, peptide, or nucleic acid) is positioned physically adja-cent to the transducer.84–89 As a result, one of the most important steps in the fabrication of a success-ful biosensor is the immobilization (or attachment) of the probe(s) onto the sensing surface.90–92 Manytypes of surfaces have been used as the substrate for biosensors, with glass and gold being examplesof the most versatile ones.35 It follows that the strategy chosen for probe immobilization depends on thetype of surface utilized.55,93Biosensors with silica-based (e.g., quartz or glass) substrates often exploit self-assembled mono-layers (SAMs) of silanes for the immobilization of the probe.94 Conversely, gold surfaces often requirea different immobilization approach.95 Gold is often used in biosensors because it is easily obtained72.1. Introduction to BiosensorsFigure 2.1: Schematic of a typical biosensor operation (left) and regeneration (right).76 Reproduced withpermission from ACS Publications.82.1. Introduction to BiosensorsFigure 2.2: Schematic of the three popular immobilization strategies for gold surfaces: physical, cova-lent, and bioaffinity immobilization.97 Modified and reproduced with permission from ACS Publications.(both as a thin film and as a colloid), widely studied, chemically inert, and electrochemically active.96 Inaddition, patterning gold on glass is easily achieved using a combination of lithographic tools (e.g., pho-tolithography and micromachining) and chemical etchants. In general, physical, covalent, and bioaffinityimmobilization strategies are popular techniques used for the attachment of probes onto gold surfaces(see Figure 2.2).97Physical immobilization is the simplest method since the probe is directly attached to the surfacewithout any chemical modification. Biomolecules (e.g., proteins) can adsorb on the surface via inter-molecular forces (i.e., mainly polar and hydrophobic interactions, and ionic bonds for proteins). However,this type of immobilization often yields a surface with weakly attached and randomly oriented probes.Alternatively, covalent immobilization often involves (1) modifying the surface with a SAM to formthe desired terminal functional group(s), (2) activating the functional group(s) on the surface, and (3)covalently coupling the probe onto the surface (please see Section 2.1.1.1 for an example of a typicalSAM for Au). This multi-step approach yields sensing surfaces with high stability suitable for prolongeduse. However, it is also difficult to control the orientation of the probes using this method.Lastly, bioaffinity immobilization offers the ability to control the orientation of the probes. In particular,the specificity between biotin and avidin/streptavidin has been widely exploited, since their interactionproduces one of the strongest non-covalent bonds known in nature. Streptavidin is a tetrameric protein(i.e., it can interact with up to four biotins), is soluble in aqueous solutions, and is stable over wide tem-92.1. Introduction to Biosensorsperature and pH ranges. Bioaffinity immobilization often involves the physical or covalent immobilizationof streptavidin onto the surface, and biotinylated probes are introduced and allowed to bind. Sometimesa biotinylated surface may be prepared to achieve a better control of the streptavidin orientation. Thestreptavidin-biotin bond formed is highly resistant to organic solvents, detergents, denaturing agents,and extreme pH and temperatures. However, chemical modification of the probe with biotin moleculesis required, which may be time-consuming or expensive. Nevertheless, this technique creates a strongand stable probe attachment with a better control over the probe orientation.2.1.1 Chemical Surface Modification Using SAMsThe type of immobilization method chosen will affect the biosensor performance (i.e. accuracy, sen-sitivity, selectivity, and stability), due to changes in the probe’s coverage and orientation.98 Forming areproducible SAM of high quality on a substrate is an active area of research since covalent immobiliza-tion of the probe is preferred over physical immobilization. Covalent immobilization may also be a criticalcomponent of bioaffinity immobilization. In this section, chemical surface modification of gold and glasssurfaces by self-assembled monolayers (SAMs) will be discussed.2.1.1.1 Thiols for Gold ModificationThemolecule used tomake a SAM typically has three key components: head group, spacer, and terminalfunctional group. Figure 2.3 shows a schematic diagram of an ideal SAM on a gold surface.Figure 2.3: Schematic diagram of an ideal SAM formed on a gold surface. The structure and character-istics of the SAM are highlighted.96 Reproduced with permission from ACS Publications.One of the most frequently used thiols for creating a SAM on gold is 11-mercaptoundecanoic acid102.1. Introduction to Biosensors(11-MUA). For 11-MUA, the head group is a thiol group, which can form a strong gold−sulfur (Au−S)bond. The spacer is a chain of saturated hydrocarbons (decane). The terminal functional group is acarboxylic acid (COOH). The 11-MUA SAM on Au can then be used to covalently immobilize probescontaining primary amino groups.99 Thermodynamic studies reveal that the energy of Au−S bond is∼15 to 30 kcal/mol (or ∼62.8 to 125.5 kJ/mol) for solution and gas-phase studies, respectively. A bondenergy of 50 kcal/mol (209.2 kJ/mol) has also been reported.96 Therefore, the adsorption of moleculeswith thiol groups onto Au is usually considered chemisorption.The self-assembly process can be performed in either a liquid environment or the gas-phase. Figure2.4 shows the simplest model of this process: (i) initial physisorption, (ii) chemisorption of the molecules,(iii) nucleation of the upright phase, and (iv) formation of highly ordered layer.100 The most commonstrategy is to immerse a clean substrate in a dilute (∼1−10 mM) ethanolic solution of thiols for ∼12−18h at room temperature. A highly packed layer can be obtained within a few minutes, but more time isrequired to allow reorganization and formation of a layer with minimal defects.96 The mechanism of SAMformation is complex, and different experimental conditions (e.g., solvent, temperature, concentration,time, purity, and chain length) can influence the structure of the final SAM.Figure 2.4: The simplest scheme of forming a SAM on Au: (i) initial physisorption, (ii) chemisorption ofthe molecules, (iii) nucleation of the standing up phase, and (iv) formation of highly ordered layer.100Modified and reproduced with permission from RSC Publishing.2.1.1.2 Silanes for Glass ModificationThe term silane (SiH4) has also been used to describe many of its derivatives. A list of commonlyencountered functional groups on silicon is shown in Table 2.1.101112.1. Introduction to BiosensorsTable 2.1: A list of common functional groups on silicon.Name Chemical GroupSilanol Si−OHSilylamine Si−NH2Methoxysilane Si−OCH3Ethoxysilane Si−OCH2CH3Chlorosilane Si−ClDisilane Si−SiDisiloxane Si−O−SiMost functionalization methods used for glass involve silanes. After creating hydroxyl groups on theglass surface (more details in Section 2.1.2), the reaction with silanes typically involves four steps. Thereaction of a methoxysilane with a surface containing hydroxyl groups is shown as an example in Figure2.5. First, the methoxy groups (OCH3) of the silane hydrolyze quickly in water environments to formreactive silanol groups. Then the silanol groups can condense with each other to form stable siloxane(Si−O−Si) bonds. Lastly, hydrogen bonding occurs between the silane and the hydroxyl groups of thesurface, leading to subsequent covalent bond formation and the release of water molecules.101,102 Itis important to note that silanes have also been observed to chemically modify many other types ofsurfaces,103 such as metal oxide94,104 and metal hydroxyl102 surfaces.Figure 2.5: The reaction mechanism of (A) hydrolysis, (B) condensation and (C, D) bonding of silanesto a hydroxyl-containing surface.102 Modified and reproduced with permission from Elsevier.122.1. Introduction to BiosensorsExperimental factors (e.g., temperature, type of solvent, water content, sample purity, immersiontime, and pH) will influence the reaction of silanes.105,106 In particular, acidic conditions catalyze thehydrolysis of silanes and reduce the self-condensation among the silanol groups. In contrast, underbasic conditions, self-condensation occurs immediately after the hydrolysis reaction, which leads to thegrowth of three-dimensional molecular structures.1072.1.2 Polydimethylsiloxane (PDMS) and BiosensorsRecently, microfluidics has been frequently combined with biosensors to extend their capabilities. Ad-vantages of microfluidics include decreased reagent volumes, increased sample throughput, decreasedreaction time, and potential for automation.108 Typically, microfluidic devices are constructed with ma-terials that form enclosed channels and cell chambers.Due to advances in the microfabrication technologies in the semiconductor industry, first-generationdevices were prepared with silicon wafers or glass.22 Glass is amorphous, electrically insulating, andoptically transparent. Silicon wafers, on the other hand, are crystalline and opaque. Both of these mate-rials were frequently used because of their resistance to organic solvents and high thermal conductivity.However, the hardness of these substrates limited their broad use. Some of the other challenges asso-ciated with silicon and glass include difficulty in bonding (e.g., high temperature and high pressure arenormally required) and high cost of fabrication (e.g., clean room environment and dangerous chemicalssuch as hydrogen fluoride are involved, thus requiring expensive protective and waste disposal facilities).Furthermore, these substrates are not permeable to gas. Consequently, they cannot be used to culturecells. These challenges motivated the search for alternative materials for microfluidics. Various poly-mers were investigated, in particular elastomeric polymers, and polydimethylsiloxane (PDMS) emergedas the most popular elastomer used in microfluidics.22 The chemical structure of PDMS is shown inFigure 2.6.Figure 2.6: The chemical structure of PDMS.109 Reproduced with permission from Elsevier.PDMS is an excellent material for creating microfluidic devices because it cures at low temperatures,132.1. Introduction to Biosensorsit is flexible, it is optically transparent, and it is non-toxic to biomolecules. Small channels can also berepeatedly reproduced by soft lithography using a mold with the desired features (see Figure 2.7). Dueto its flexibility, PDMS can be easily removed from the mold. Subsequently, it can form reversible bondsto other materials via intermolecular (van der Waals) forces, or it can form irreversible bonds to glass orother PDMS after exposure to air or oxygen plasma by forming siloxane bonds.110,111Figure 2.7: Scheme for producing PDMS chips using soft lithography. (A) A mold is made with thedesired features. (B) Liquid pre-polymer material is poured onto the mold. (C) Mold with liquid PDMSis cured at low temperatures. (D) Solidified and flexible PDMS is peeled off and (E) placed on a rigidsubstrate for experiments.1 Reproduced with permission from ACS Publications.The surface of unmodified PDMS is hydrophobic. As a result, hydrophobic PDMS channels aresusceptible to bubble formation and non-specific adsorption of hydrophobic molecules. One way tomitigate these challenges is via surface functionalization. Most of the functionalization methods appliedto PDMS use silanes. This strategy is analogous to the standard glass-based surface chemistry aspreviously described.94 Prior to silanization, the surface is first treated to remove organic residues andto increase the number of hydroxyl (−OH) groups on the surface (i.e., silanols Si−OH). The most widelyused method for PDMS (and glass) is using oxygen plasma.111,112 The literature also describes othermethods such as exposing to UV-ozone and soaking in sodium hydroxide.94,113,114 This cleaning andoxidizing step is important for the effective silane functionalization.142.1. Introduction to Biosensors2.1.3 An Example: Surface Plasmon Resonance (SPR) BiosensorSurface plasmon resonance (SPR) is a powerful and versatile optical technique commonly used for an-alyzing biomolecular interactions.115 Typically, a glass substrate coated with a thin Au film is used forSPR applications. At a specific angle of incidence (i.e., resonance angle), a reduction in the intensity ofthe reflected light can be detected.116 When the refractive index of the interfacial region changes (e.g.,due to protein adsorption on Au), the angle of incidence required for the resonance will change as well(see Figure 2.8).117 Therefore, by monitoring the change in the resonance angle (or reflected light inten-sity) as a function of time, the interaction between the analyte molecules (i.e., targets flowing in solution)and the biorecognition elements (i.e., probes immobilized on the gold substrate) can be examined andanalyzed.118 SPR biosensors are commonly used to obtain binding kinetics and equilibrium constantsfor biomolecules,119 due to its inherent surface sensitivity and its virtually real-time response.120Figure 2.8: A typical setup of an SPR biosensor.118 Reproduced with permission fromNature PublishingGroup.152.1. Introduction to Biosensors2.1.3.1 Carbodiimide Activation ChemistryCovalent immobilization is one of the most frequently employed strategies for the attachment of probesonto the gold surface used in SPR biosensors, especially via the carboxyl-amine coupling chemistry.121Figure 2.9 shows a typical carboxyl-amine coupling mechanism using the carbodiimide reaction chem-istry. First, the gold surface is modified either with a SAM or layer of dextran containing carboxyl groups.Then a mixture of N-hydroxysuccinimide (NHS) and N-(3-(dimethylamino)propyl)-N´-ethylcarbodiimidehydrochloride (EDC) solution is injected over the surface to activate the carboxyl groups.122 Subse-quently, the covalent coupling of biomolecules containing primary amino groups to the surface is achievedby forming amide bonds. The unreacted succinimide groups are usually quenched with ethanolamine.48,121,123,124Figure 2.9: The chemical reaction scheme for the covalent immobilization of a probe onto the surfacevia carboxyl-amine coupling chemistry. (1) The carboxyl group is activated with EDC/NHS. (2) Covalentattachment of a probe by its primary amino group. (3) The unreacted active site can be deactivated withethanolamine.121 Modified and reproduced with permission from Springer.2.1.3.2 PDMS-Based SPR BiosensorMost of the commercially available SPR systems, such as BIAcore 3000, have difficulties in perform-ing high-throughput assays, concentration assays, and multiple interactions.125 Much effort has beendevoted to address these issues by combining microfluidics with SPR imaging (SPRi) techniques to162.1. Introduction to Biosensorsdecrease reaction time, reduce reagent consumption, allow concentration analysis, and increase assaythroughput.126–128 In SPRi, usually a microarray of gold spots is formed on the glass substrate in orderto create high-contrast SPR images, minimize cross-talk between adjacent sensing channels, and re-duce background noise.129–131 SPRi usually measures at a fixed angle, where differences in reflectivityobserved at the array interface are monitored over time with a camera. Microfluidic SPRi allows for mul-tiplexed detection for high-throughput bio-analysis.132 Figure 2.10 shows the design of a PDMS-basedmicrofluidic device for SPRi applications.18Figure 2.10: The design of a PDMS-based microfluidic device for SPRi applications.18 Reproduced withpermission from RSC Publishing.The PDMS microchip shown in Figure 2.10 could simultaneously monitor multiple analyte streamsagainst different probes by using a parallel 264-chamber microarray. The addition of a dilution networkand a system of valves and pumps illustrates the “lab on a chip” concept. Importantly, due to designcomplexity and the high pressure required, the PDMS microchip should be strongly bonded to the glasssubstrate so that the biomolecular interaction analysis could be performed on the Au spots. Modifyingthe Au spots with a SAM was not possible because of the harsh UV-Ozone cleaning/bonding method.172.2. Surface AnalysisTherefore, physical immobilization of the proteins was utilized. However, the weakly and randomly ad-sorbed proteins could be easily washed off by the running buffer,97 thus affecting the SPR curves (aswell as the binding kinetics and equilibrium constants) obtained.2.2 Surface AnalysisThe analysis of surfaces modified with self-assembled monolayers (SAMs) is an active area of research,since the characteristics of the adsorbed layer can affect the device performance. To better understandthe process of SAM formation and the physical/chemical properties of the adsorbed layer, a wide rangeof surface-specific analysis techniques have been utilized.133 In the following sections, the techniquesemployed in this thesis are introduced.2.2.1 Water Contact AngleIn nature, we observe that a small drop of water typically forms a somewhat spheroidal shape, sincethis shape has the lowest surface area for a given volume of water. This tendency to minimize surfacecontact area is characterized as surface tension γ. Surface tension is defined as a force per unit length(or energy per unit area) used to create a new surface.When a drop of liquid is placed on a solid surface, a triple interface is formed between the solid, liquidand gas as shown in Figure 2.11, where γSL is the surface tension of the solid-liquid interface, γLG isthe surface tension of the liquid-gas interface, and γSG is the surface tension of the solid-gas interface.The angle between the solid surface and the tangent to the liquid surface is known as the contact angleθC (measured in the liquid phase).134Figure 2.11: Illustration of contact angles formed by sessile liquid drop on a solid surface.134 Modifiedand reproduced with permission from Springer.182.2. Surface AnalysisAt equilibrium, these tensions will be in balance according to Young’s Equation as shown eq 2.1:γSG = γSL+ γLGcos(θC) (2.1)It is important to note that cohesive forces exist between molecules in the liquid drop, and adhesiveforces exist between the liquid molecules and the surface. For example, if there are polar groups on thesurface (e.g., hydroxyl groups) and a water drop is used, there will be strong adhesive forces betweenthe water molecules and the surface. This type of a surface is called hydrophilic and a low contact angleis observed (as shown by Figure 2.11 on the left). Conversely, if the surface consists mainly of non-polargroups (e.g., surface is covered with an organic layer), the surface is hydrophobic and a large contactangle is formed (as shown by Figure 2.11 on the right). As a result, contact angle measurements areused as a quick and simple technique to obtain qualitative information about the chemical nature of asurface.The sessile drop method is commonly used, where measurements of the shape of a liquid dropsitting on a flat surface are made with a camera. Software can be used to analyze the digital images todetermine the contact angle. Wettability is determined by the equilibrium contact angle θC. If θC < 90°,the liquid is said to wet the solid (complete wetting occurs when θC = 0°). If θC > 90°, the liquid doesnot wet the solid. In particular, the sessile drop method has been applied to study mixed SAMs ofalkanethiols on gold135 and the hydrophobic recovery of PDMS136 by oxygen plasma and chemicaltreatment.2.2.2 Atomic Force Microscopy (AFM)Atomic force microscopy (AFM) is one of the most frequently used tools for imaging surface topographyand measuring certain physical properties (i.e., hardness, friction, and thickness) of the surface withatomic-scale resolution. A diagram of the typical AFM setup is shown in Figure 2.12. A laser lightsource (with the help of some focusing optics) is reflected off the back of a cantilever, which has a sharptip at the end. The small sharp tip used usually has a radius of ∼ 10 to 100 nm.133 When the tip movesclose to the sample surface, the tip−surface interaction can cause a deflection of the cantilever, whichcan be measured as a change in the deflected light and recorded by a detector.Numerous AFM operation modes have been proposed to image the surface.137 As the cantilever israster scanned across the sample, the topography of the sample is most commonly measured in eitherthe constant-deflection or the constant-height mode.138,139 In the constant-deflection mode, the can-192.2. Surface Analysistilever deflection is kept constant by extending and retracting the piezoelectric scanner. In the constant-height mode, the fixed end of the cantilever is kept at a constant height and the cantilever tip deflectiondue to sample-tip interaction is recorded. In this thesis, contact-mode imaging (i.e., imaging mode andlateral force mode) and force-distance curve were used.Figure 2.12: Schematic diagram showing the setup of a typical AFM experiment.138 Reproduced withpermission from Elsevier.2.2.2.1 AFM Contact-Mode ImagingIn this imaging mode, the vertical displacement of the cantilever is controlled to keep a constant deflec-tion of the probe as the tip scans laterally on the sample surface. This mode can be operated in eitherthe attractive or repulsive regime. The variation of the displacement as a function of lateral position givesinformation about the surface topography. Lateral force (or friction force) mode is similar to the imagingmode. However, during the lateral scan of the tip over the surface, the twisting of the cantilever (due tothe force between the tip and the surface) is also monitored. These techniques have been applied tostudy patterned SAM formation on Au.140202.2. Surface Analysis2.2.2.2 AFM Force-Distance CurveIn a force-distance curve measurement, the displacement between the sample and the fixed end of thecantilever is changed, while the attractive (or adhesive) and repulsive forces between the sample andthe cantilever tip are monitored (see Figure 2.13). At position A, the sample surface is far away from thecantilever tip, so the force is zero. As the tip is moved closer to the sample, the attractive force betweenthe tip and the sample surface begins to pull the tip downward. At position B, the tip is in contact withthe sample surface. As additional force is applied to the cantilever, the sample-tip interaction becomesrepulsive as shown by position C. The tip is then retracted with the tip and sample still in contact throughsome range of force, as shown by position D. Finally, at a distance beyond the maximum adhesive force,the tip snaps out of contact with the surface (position E) andmoves away from the sample (position F).141Figure 2.13: A schematic of force-distance curve.141 Reproduced with permission from Journal of CellScience and Company of Biologists LTD.The net interaction between the sample and the tip results from a sum of different forces. Beforecontact, long-range interactions (i.e., van der Waals and Coulomb forces) exist. Once in contact, chem-ical bonds may form between the tip and the surface. As the tip is removed from the sample, the pull-offforce gives a measure of the adhesion force. The AFM force-distance curve has been applied to studythe surface acid−base properties of SAMs with terminal carboxylic acid groups.142212.2. Surface Analysis2.2.3 Infrared (IR) SpectroscopyInfrared (IR) spectroscopy uses the fact that a molecule absorbs light at different frequencies in the IRregion that are characteristic of its chemical bonding configuration and three-dimensional structure. ForIR radiation to be absorbed, the molecule’s electric dipole moment must change (e.g., vibrations androtations). Vibrations can involve a change either in bond angle (bending) or in bond length (stretching),such as symmetrical stretching or asymmetrical stretching. Even for a simple molecule, many differenttypes of vibrations are observed.143Two methods for measuring IR spectra will be described: attenuated total reflectance Fourier trans-form infrared spectroscopy (ATR-FTIR) and polarization modulation-infrared reflection-absorption spec-troscopy (PM-IRRAS). Some applications of IR spectroscopy will be highlighted in Section 2.2.5.2.2.3.1 ATR-FTIRTraditional transmission spectroscopy is a simple method for the collection of IR spectra. Prior to ex-periments, a solid sample must be finely ground and diluted with IR transparent salt (e.g., potassiumbromide) and pressed into a thin film (or pellet). As the beam passes through the sample, some ofthe light is absorbed and this method measures the percentage of IR light transmitted (or % Transmit-tance) at specific wavelengths. Liquid, solid, or gas samples can be analyzed using this approach.144In recent years, reflectance sampling techniques are becoming more popular, since they require less(or simpler) sample pre-treatment. In particular, attenuated total reflectance (ATR) spectroscopy is anon-destructive method that has been applied to study surfaces, films, and solutions. The sample maybe directly placed on the ATR crystal (with some pressure applied) and the IR spectra can be quicklyobtained (see reference for a typical setup).145ATR-FTIR employs the phenomenon of total internal reflection (TIR). TIR occurs when light is com-pletely reflected at the interface between two optically different media (i.e., with different refractive in-dices) at an angle larger than the critical angle (with respect to the normal to the surface). This processcreates an evanescent wave that travels along the boundary between the two materials. First, the IRbeam is directed at a crystal of higher refractive index (at a particular angle of incidence). An evanescentwave, created by the internal reflection(s), extends into the sample held in contact with the crystal. Inthe spectral regions where the sample absorbs energy, the evanescent wave will be attenuated. Thenthe reflected radiation (with some sample absorption) is returned to the detector.143 Many experimentalfactors can influence the measurements (e.g., number of reflections and quality of contact between the222.2. Surface Analysissample and the ATR crystal).2.2.3.2 PM-IRRASInfrared reflection-absorption spectroscopy (IRRAS) is a well-established technique frequently used tostudy monolayers and thin films deposited onto metallic surfaces.146 To overcome limitations of IRRAS(e.g., inability to detect ultra-thin films or lengthy experimental time periods), polarization modulation-IRRAS (PM-IRRAS) was introduced144. ThePM-IRRASmeasurement depends on the polarization of the in-cident IR beam, the angle of incidence, and the optical constants of the thin film and substrate (as wellas the molecule’s orientation on the surface).147Figure 2.14: Illustration of the phase change upon reflection at a metal surface for the components ofthe incident radiation polarized perpendicular (s) and parallel (p) to the plane of reflection.147 Modifiedand reproduced with permission from ACS Publications.First, IR radiation illuminates the metal surface at a well defined and controlled angle of incidence.A photoelastic modulator generates this radiation, which is either parallel (p-polarized) or perpendicular232.2. Surface Analysis(s-polarized) to the plane of incidence. When the beam is reflected from the surface, the electric vectorexperiences a phase change, whose magnitude depends on the polarization (see Figure 2.14).At almost all angles of incidence, the radiation polarized perpendicular to the plane of incidence(i.e., s-polarized) undergoes a phase change of π and the electric vectors sum to near zero at thesample, which results in a zero electric field strength. Therefore, no IR absorption occurs (see Figure2.15). Near the grazing angle of incidence (i.e., about 88°), the radiation polarized parallel to the plane ofincidence (i.e., p-polarized) undergoes a different phase change of about π/2 and the resultant standingwave is non-zero at the surface and is oriented along the surface normal. As a result, IR absorption isenhanced.144,147Figure 2.15: Dependence of the absorption factor at the wavelength of maximum absorption on θ for a1 nm layer of acetone on metal.147 Reproduced with permission from ACS Publications.It follows that the dipole transition moment of the adsorbed molecule(s) on the surface must have acomponent oriented along the surface normal in order to absorb the incident radiation. As a result, onlythe parallel component of reflectivity (Rp) will be influenced by these surface adsorbed molecule(s). In242.2. Surface Analysiscontrast, the randomly oriented molecules in the beam path (e.g., water vapor and carbon dioxide in air)will influence both Rp and Rs.With a single detector, the sum of the reflectivity of the p- and s-components (Rp+ Rs) and thedifference in the reflectivity (Rp − Rs) may be obtained using a photoelastic modulator and a lock-inamplifier. The PM-IRRAS signal is given by the differential reflectivityΔRR=(Rp−Rs)(Rp+Rs), which is proportionalto the absorbance. Therefore, the final PM-IRRAS spectrum contains signals/absorbance from only thesurface adsorbed species with dipoles oriented along the surface normal.2.2.4 X-ray Photoelectron Spectroscopy (XPS)X-ray photoelectron spectroscopy (XPS) operates on the principle of the photoelectric effect (see refer-ence for more information).148 Illumination of a surface with monochromatic X-ray radiation results in aprimary excitation process that produces electrons (also called photoelectrons). These electrons maybe ejected from the sample surface into the surrounding vacuum with minimal loss in kinetic energy.Therefore, the distribution of unscattered electrons as a function of their kinetic energies, which can beconverted to binding energies, results in the XPS spectrum.The energy of the overall process must be conserved and is described by eq 2.2, where h is Planckconstant, ν is frequency, Eb is binding energy, Ek is kinetic energy, Er is recoil energy, and φsp isspectrometer work function.hν = Eb+Ek +Er +φsp (2.2)The two most important of these quantities are the kinetic energy (Ek) of the electron inside thespectrometer and the energy required to remove the electron from the initial state (i.e., binding energyEb). Experimentally, the binding energy (Eb) can be approximated by hν−Ek . Because the energy ofthe X-ray (with a particular wavelength) is known, the discreteEk valuesmeasured can be correlated withthe Eb values of different atomic levels. The resulting spectrum shows a peak corresponding to eachenergy level. For accurate binding energy assignments, energy contributions from the recoil energy (Er)and the spectrometer work function (φsp) must also be considered.XPS is a surface-sensitive technique that measures the elemental composition, as well as the chem-ical and electronic states of the elements that exist on the sample surface. Quantitative information canalso be obtained by counting the number of electrons that escape from the sample surface.252.2. Surface Analysis2.2.5 Surface Analysis of Silanes on GoldOne of the main goals of this thesis is to characterize the adsorption of a carboxylated silane (TMS-EDTA) on Au in order to better understand the physicochemical nature of this interface and therebydetermine the properties relevant to its application to microfluidic electrochemical devices. Therefore,the literature related to the surface analysis of silanes on gold will be presented here. The adsorp-tion of chlorosilane149 and mixed monolayers of silanes on gold,150 and the interaction of silanes withother oxide-free metal151 substrates have been reported. In the following sections, the use of some ofthe surface analysis techniques described above to investigate the adsorption of silanes on Au will behighlighted. Specifically, the adsorption of an ethoxysilane and alkylsilanes on Au will be discussed.2.2.5.1 Amino-Silane for Gold ModificationThin films of (3-aminopropyl)triethoxysilane (3-APTES) on aluminum oxide and planar gold substrateshave been analyzed by contact angle measurements and IR techniques (see Figure 2.16).152 Contactangle measurement of 3-APTES on aluminum substrate showed a water contact angle of 65°. Thewetting behavior was a result of several contributions, including amino, ethoxy, and silanol groups (aswell as a high degree of siloxane cross-linking).Figure 2.16: Reflection-absorption Fourier transform infrared spectroscopy (RAIR) spectra of 3-APTESadsorbed on gold from methanol solution, as a function of hydrolysis and condensation reactions. (1)Freshly deposited 3-APTES at room temperature, (2) plus 4 min at 75 °C, (3) plus 6 min, (4) plus 6 min,(5) plus 20 min, and (6) after heating for 12 h at 75 °C.152 Modified and reproduced with permissionfrom ACS Publications.Prior to modifying the gold substrate, normalized transmission IR spectra of 3-APTES were ob-tained to help with the assignment of band positions and modes. Subsequently, reflection-absorption262.2. Surface AnalysisFourier transform infrared spectroscopy (RAIR) was used to study the hydrolysis and condensation of3-APTES on gold. Figure 2.16 shows the absorption IR spectra for two regions: the alkyl stretchingregion (3100–2700 cm−1) and the fingerprint region (1800–900 cm−1).Freshly deposited 3-APTES (spectrum 1) shows Si−O modes with maxima at 1126 and 1091 cm−1and some bands from the ethoxy groups (e.g., 2975 and 1390 cm−1). The Si−O modes broaden anddevelop several maxima as the silane cross-links (spectrum 5). At the same time, the ethoxy modesdecrease in intensity (e.g., at 2974 cm−1). For spectrum 6, the film is extensively cross-linked (i.e., theSi−O region shows one major peak at 1150 cm−1 and a shoulder at 1050 cm−1). Evidently, hydrolysisand condensation of the silane (i.e., siloxane cross-linking) were observed.2.2.5.2 Alkylsilanes for Gold ModificationRecently, the adsorption of silane (SiH4) and methylsilane (CH3−SiH3) on gold has been studied.67Furthermore, several alkylsilanes with longer hydrocarbon chains have also been studied by combin-ing IR and XPS techniques.153–155 For the experiments with alkylsilanes, Figure 2.17 shows the IRtransmission spectra for 3050–2750 and 2350–2000 cm−1 regions (the latter region can analyze thesilicon-hydrogen stretch at 2150 cm−1).Figure 2.17: RAIR spectra for monolayers of (A) octadecylsilane, (B) octylsilane, and (C) hexylsilane onAu. A solution IR spectrum of octylsilane (D) is included for comparison.153 Reproduced with permissionfrom ACS Publications.The observed carbon-hydrogen stretching modes between 2850 and 3000 cm−1 are consistent withthe alkyl chains. The silicon-hydrogen stretch at 2150 cm−1 is observed for the liquid alkylsilane (Figure272.2. Surface Analysis2.17D). However, this feature is absent in the other spectra. These results suggest that no silicon-hydrogen bonds remain after chemical adsorption.Figure 2.18 shows the XPS spectra for C 1s electronic energy level (-288.5 to -278.5 eV), as wellas Si 2p and Au 4f (-103 to -83 eV) core-levels. When compared to the Au reference (spectrum 4), allsamples (spectra 1-3) show an increase in the C atoms (Figure 2.18A) and an increase in the Si atoms(Figure 2.18B).Figure 2.18: XPS spectra for (A) C 1s core-levels, and (B) Si 2p and Au 4f core-levels of (1) octadecyl-silane, (2) octylsilane, and (3) hexylsilane chemisorbed on (4) freshly evaporated Au.153 Modified andreproduced with permission from ACS Publications.The combination of XPS and IR data showed that all three silicon-hydrogen bonds reacted with thegold surface, and silicon formed bonds with three gold atoms on the surface. The proposed mechanismof alkylsilane adsorption on Au is shown in Figure 2.19.Figure 2.19: The proposed reaction mechanism of a monolayer of silane on Au.153 Reproduced withpermission from ACS Publications.Subsequently, the oxidation of alkylsilane-based monolayers on gold was studied.155 XPS spectra282.3. Thermodynamics of Adsorptionof freshly prepared samples and samples treated with O3/O2are shown in Figure 2.20. Exposing thefreshly prepared sample (I) to ozone results in the oxidation of the alkyl chain. For example, there isan increase in O (Figure 2.20A) and loss in C (Figure 2.20B). Furthermore, the increasing shoulder at∼ −287 eV represents the formation of carboxyl groups in the monolayer.Figure 2.20: (A) O 1s and (B) C 1s core levels of an octylsilanemonolayer on gold exposed to successivedose of O3/O2.155 Reproduced with permission from ACS Publications.These studies indicate that some silanes can be used to chemically modify Au surfaces. Further-more, different silanes result in differing surface adsorbed structures. For example, 3-APTES modifiesthe Au surface with extensive siloxane cross-linking, forming multilayers. On the other hand, alkylsi-lanes adsorb onto the Au surface with little evidence of cross-linking, most likely forming monolayers.The combined use of the surface analysis techniques described above will help to elucidate the chemicaland physical nature of the adsorbed TMS-EDTA layer on Au.2.3 Thermodynamics of AdsorptionGenerally, adsorption deals with the physical and/or chemical interaction of molecules, atoms, or ions(of gas, liquid, or dissolved solid) with a surface,156 and this process creates a layer of adsorbate con-tacting the surface. Solvent-based chemical modification is often employed to prepare a surface withself-assembled monolayer (SAM). As a result, the adsorption of a molecule for creating the SAM is incompetition with solvent and other species in the solution. It is important to gain an understanding of the292.3. Thermodynamics of Adsorptionmolecule’s Gibbs free energy of adsorption,157,158 since this information will in turn provide an indica-tion of the process spontaneity. The change in Gibbs free energy depends on the adsorbate/adsorbentbond strength (enthalpy change) and entropy change.159 Depending on this strength, the adsorptionprocess can be classified either as physisorption (e.g., weak van der Waals forces) or chemisorption(e.g., covalent bonding). For some species, adsorption may also occur due to electrostatic attraction.160Moreover, an understanding of the molecule’s adsorption process can potentially lead to the design ofimproved biosensors and devices,96 since complex biological samples are often analyzed and multipleuse of the devices may be required.2.3.1 The Gibbs Adsorption IsothermFor biosensors, the solid substrate (e.g., glass or gold) is often in contact with an aqueous solution. Theinterphase of such a system can be divided into three regions: two distinct bulk phases (i.e., solid andliquid) and a surface phase defined as the interphase region. Experimentally, it is difficult to determinethe exact structure of this surface phase.Josiah Willard Gibbs proposed an idealized model that defined the interphase region as having zerothickness. In this model, the chemical components of the two bulk phases remain unchanged exceptnear the dividing surface. For a particular species , the quantitative measure of adsorption at thedividing surface is captured by the surface excess quantity () with respect to bulk (i.e., an arbitraryplane defined in the bulk solution) as shown in eq 2.3, where n is the number of moles in excess andA is the area of the dividing surface. = n/A (2.3)The Gibbs adsorption isotherm, given by eq 2.4, provides the simplest description of the propertiesof interphase. In this equation, γ is the surface tension, and  is the surface excess and μ is thechemical potential for a particular species .− dγ=∑dμ (2.4)By definition, surface tension is equivalent to the Gibbs free energy per unit area of interface.133Equation 2.4 suggests that, for systems not at equilibrium, there is a natural tendency for the Gibbsenergy at constant temperature and pressure to decrease. For example, a pure phase always assumes302.3. Thermodynamics of Adsorptiona shape that creates the minimum surface area per unit volume. In addition, when a solution is incontact with another phase, the composition of the interphase differs from that of the bulk such that dγ isminimized at constant temperature and pressure. From eq 2.4, it is evident that when the surface excessof a species is positive (> 0), increasing the chemical potential of that species (e.g., by increasing itsconcentration in the bulk liquid phase) decreases the surface tension (dγ < 0). Conversely, when thesurface excess of a species is negative, increasing the chemical potential of that species increases thesurface tension. For examples, surfactants have a positive surface excess concentration and induce adecrease in surface tension; conversely, electrolytes have a negative surface excess concentration, andinduce an increase in surface tension.2.3.2 Langmuir Adsorption IsothermThe interphase region (or surface phase) of a real system is different from the one proposed by Gibbs.In particular, if we assume that the molecules can only adsorb as a monolayer on the solid surface, thenthe surface coverage (θ) can be defined as shown in eq 2.5 at equilibrium, where  is the surface excessof a particular species and m is the surface excess at maximum surface coverage. This concept isfundamental to the derivation of one of the simplest models for adsorption: the Langmuir adsorptionisotherm. It is important to keep in mind that both Langmuir and Gibbs adsorption isotherms are idealmodels.θ=m(2.5)The Langmuir adsorption isotherm relates the surface coverage of a particular species (θ) to its bulkconcentration (C). This model assumes that the adsorbed layer is a monolayer and the species do notinteract with each other. Furthermore, there is a one-to-one relationship between the species and theadsorption sites (and the free energies of all adsorption sites are assumed to be equivalent).161 TheLangmuir adsorption process can be represented by eq 2.6, where A represents the free molecules (oradsorbates) that can adsorb onto the surface, S represents the free adsorption sites on the surface, andAd represents the adsorbed molecules (or adsorbates) on the surface.A+ S⇌ Ad (2.6)Using the definition that (1− θ) =Nƒ reeNs(where Nƒ ree is the number of free sites and Ns is the totalnumber of sites), the rate of adsorption (r) can be written as312.3. Thermodynamics of Adsorptionr = k[A][S] = k[A]NS (1− θ) (2.7)and the rate of desorption (rd) can be written asrd = kd[Ad] = kdNSθ (2.8)(where k and kd are the rate constants for adsorption and desorption, respectively). The dynamicequilibrium process can be represented by eq 2.9.k[A]NS (1− θ) = kdNSθ (2.9)Let us represent [A] as C (i.e., the bulk concentration of the adsorbing molecule), we get eq 2.10.kC(1− θ) = kdθ (2.10)If we define the Langmuir equilibrium constant as KL (= k/kd), then we get eq 2.11.KL =θC(1− θ)(2.11)Solving for θ, we get eq 2.12 (i.e., the Langmuir adsorption isotherm).θ=KLC1+ KLC(2.12)2.3.3 Standard Gibbs Free Energy of Adsorption ΔG°Adsorption is typically a spontaneous process, and is indicated by the negative Gibbs free energychange. The standard Gibbs free energy (ΔG°) can be calculated at equilibrium using eq 2.13 un-der standard conditions, where R is the gas constant, T is the temperature in Kelvins, and K is thethermodynamic equilibrium constant.ΔG°=−RT lnK (2.13)Experimentally, by changing the bulk concentration C, different values for surface coverage θ canbe obtained. Then, the concentration-dependent surface coverage can be fit to Langmuir adsorption322.3. Thermodynamics of Adsorptionisotherm (eq 2.12) to obtain the equilibrium constant KL using non-linear least squares fitting. Conse-quently, the Langmuir adsorption free energy (ΔG◦ds) can be obtained.2.3.4 Frumkin Adsorption IsothermOne major drawback of the Langmuir adsorption isotherm is that it does not take into account the inter-action among the adsorbed molecules. For instance, when negatively charged species are adsorbed onthe surface, theremay be electrostatic repulsion among the chargedmolecules. The Frumkin adsorptionisotherm was developed to take these interactions into account.The improved adsorption isotherm is shown in eq 2.14, where KF is the Frumkin equilibrium constant,and α is the Frumkin lateral interaction parameter.θ=KFCexp(−αθ)1+ KFCexp(−αθ)(2.14)The neighbouring molecules may interact due to either attraction or repulsion. According to eq 2.14,positive α values indicate repulsion between the neighboring molecules. Negative α values indicateattraction between the neighboring molecules. When α= 0, the Frumkin isotherm becomes identical tothe Langmuir isotherm.The concentration-dependent surface coverage can also be fit to Frumkin adsorption isotherm (eq2.14) to obtain both KF and α. The Frumkin ΔG◦dscan also be obtained (using eq 2.13).2.3.5 ΔG◦dsin a Complex EnvironmentThere are also many other adsorption isotherms proposed (e.g., Henry isotherm for low surface cover-ages and Temkin isotherm for uncharged and non-interacting species),159,162 but they do not adequatelyexplain the complex TMS-EDTA system. Therefore, their discussion is beyond the scope of this thesis.Historically, the adsorption isotherms have been derived for gas−solid systems. However, the equa-tions derived above are also valid for liquid−solid systems, under specific conditions. It is importantto note that for studies performed under aqueous conditions, water and components of the electrolytemay also adsorb onto the surface. As a result, the Langmuir equilibrium constant (KL) obtained for aparticular species may be affected by the displacement of water molecules and electrolyte. Similarly,the Frumkin equilibrium constant (KF) and the lateral interaction parameter (α) obtained for a particularspecies (under aqueous conditions) may also be affected. Therefore, the free energy of adsorption,ΔG◦ds, is uniquely specified for a particular liquid−solid interphase.332.4. Electrochemistry2.4 ElectrochemistryIn this thesis, electrochemical techniques have been applied to study TMS-EDTA adsorption on Au underalkaline aqueous conditions, in the presence of chloride ions. Electrochemical differential capacitance isone of the best techniques to determine the standard free energies of adsorption in a complex aqueousenvironment, while also elucidating its dependence on the potential of the metal substrate. Therefore,the background theory of electrochemistry will be reviewed in the following sections. First, the elec-trode−solution interface, with an emphasis on the solution side of the interface, will be discussed.2.4.1 Electrical Double LayerTypically, electrochemistry involves the use of a metal (i.e., electrode surface) in contact with an aqueoussolution containing a high concentration of ions. When excess charge (i.e., excess or deficiency ofelectrons) accumulates as a thin layer at the metallic side of the interface (assuming that the metal is agood electrical conductor), the counter-charge in solution is made up of an excess of either cations oranions near the electrode surface. The distribution of charged species and oriented dipoles that exists atthe solution side of the interface is called the electrical double layer (or double layer).163 Several modelsof the double layer have been proposed and its historical development will be described below.2.4.1.1 Helmholtz ModelIn 1853, Hermann von Helmholtz proposed that charged electrodes (immersed in an aqueous solutionof electrolytes) repel the ions of similar charge and attract the ions of dissimilar charge to their surfaces.As a result, two layers of opposite polarity (separated by a distance of molecular order) form at themetal−solution interface. This model is equivalent to a parallel-plate capacitor, which can be representedby eq 2.15, where σ is the stored charge density, ϵ is the dielectric constant (or relative permittivity) ofthe medium, ϵ0 is the permittivity of free space, d is the spacing between the two plates, and V is thevoltage drop.σ =ϵϵ0dV (2.15)This model predicts that the potential profile changes linearly with distance. Also, the differentialcapacitance per unit surface area (Cd) can be derived as342.4. Electrochemistry∂σ∂V= Cd =ϵϵ0d(2.16)This model suggests that Cd is a constant; however, this result is not typically observed. Experimen-tally, the Cd values depend on both electrode potential and electrolyte concentration. Therefore, theHelmholtz model was revised.2.4.1.2 Gouy-Chapman ModelLouis Georges Gouy in 1910 and David Leonard Chapman in 1913 independently proposed modifica-tions to the Helmholtz model. Because the conductivities of the two sides of the capacitor are distinctlydifferent, the Gouy-Chapman model suggests that a thicker layer of charge on the solution side of thisdouble layer is required. This layer is known as the diffuse layer of charge, and it extends severalnanometers into the solution.These ions in solution interact electrostatically with the excess surface charge on the surface ofthe metal. These ions also move in the solution due to thermal motion. In this case, the potentialprofile changes non-linearly with distance. Evidently, an average distance of charge separation (orcharacteristic length) needs to replace d in eq 2.16. This characteristic length also varies with electrodepotential and electrolyte concentration. For example, when the electrode becomes more highly charged(or when the electrolyte concentration rises), the diffuse layer becomes more compact and Cd also rises.This model works well at low concentrations and at potentials near the point of zero charge (PZC),where the charge at themetal in contact with the electrolyte equals zero (i.e., if the charge of the electrodesurface is zero, then there is no accumulation of oppositely-charged ions on the solution side). However,it fails for extreme potentials and high electrolyte concentrations. Hence, the Gouy-Chapmanmodel alsowas revised.2.4.1.3 Stern ModelOne of the deficiencies of the Gouy-Chapman model is that the ions are assumed to be point charges.However, this view is not realistic because an ion has a finite size and cannot approach the surface anycloser than the ionic radius. In 1924, Otto Stern suggested combining the Helmholtz model with theGouy-Chapman model. This new model proposes that some ions adhere to the electrode as suggestedby Helmholtz (called the Stern or Helmholtz layer), while some ions form a Gouy-Chapman layer (calledthe diffuse layer).164352.4. ElectrochemistryEquivalently, this model represents two capacitors in series: one for the Helmholtz layer (CH) andthe other one for diffuse layer (CD). Mathematically, the interfacial capacitance derived from the Gouy-Chapman-Stern model can be represented by eq 2.17, where Cd is the capacitance of the double layer(i.e., total capacitance).1Cd=1CH+1CD(2.17)2.4.1.4 The Role of the Solvent at the InterfaceThe Helmholtz/Stern layer accounts for the ion’s finite size and its ionic radius, but this model still hassome deficiencies and has been further refined (e.g., by David C. Grahame165 in 1947, and by Bock-ris/Devanathan/Muller166 in 1963). Some of the most important modifications are summarized here.Depending on the type of solvent, different ions in solution are solvated differently. The interactionbetween an ion and the surface depends on its degree of solvation. For ions that are strongly hydratedin aqueous solutions, the interaction with the surface is mainly electrostatic. On the other hand, the ionsthat are not hydrated (or partially hydrated) can be in direct contact with the electrode surface. Thesespecies are called specifically adsorbed ions.In addition, the solvent molecules, such as water, has a strong dipole and will be influenced bythe charge on the metal surface. In particular, water would act as a first solvation shell for the metalelectrode and would exist in a fixed orientation. At some sites, the water would be displaced by thespecifically adsorbed ions. Figure 2.21 shows a schematic representation of the electrical double layerof a negatively charged electrode.It is important to note that the solution side of the double layer can be divided into several layers. Theinner layer is the closest to the electrode and it contains solvent molecules and sometimes specificallyadsorbed species. This inner layer is also called the compact, Helmholtz, or Stern Layer. The innerHelmholtz plane (IHP) passes through the centres of the specifically adsorbed species at a distanceof 1. In contrast, the solvated ions can only approach the metal to a distance of 2, and the outerHelmholtz plane (OHP) passes through the centres of these solvated ions. Since the solvated ions in-teract with the charged metal via only long-range electrostatic forces, they are said to be non-specificallyadsorbed. These non-specifically adsorbed ions are distributed, due to thermal motion, in a region calledthe diffuse layer (i.e., the region between the OHP and the bulk of solution).362.4. ElectrochemistryFigure 2.21: A schematic representation of the electrical double layer of a negatively charged inter-face.167 Reproduced by permission of The Electrochemical Society.2.4.2 Adsorption on an Electrode SurfaceFrom Figure 2.21, it is evident that the specifically adsorbed species need to displace water moleculesin order to be in contact with the electrode surface. It is important to note that the sum of charges on themetal and solution side should be zero. The solvent-based adsorption of self-assembled monolayerson Au undergoes a similar displacement process. In the following sections, the theory of adsorption onan electrode surface and use of differential capacitance measurements to obtain surface coverage andfree energies of adsorption will be introduced.2.4.2.1 Electrocapillary EquationThe energies of the electrode−solution interface can be obtained by measuring its surface or interfacialtension. The surface tension of an electrode in contact with an electrolyte depends on themetal−solutionpotential difference, as shown by the electrocapillary equation (eq 2.18).168 The surface excess on372.4. Electrochemistrythe solution side is denoted by∑dμ (see Section 2.3.1). The surface excess on the metal side isexpressed by σMdE, where σM is the excess surface charge density of the metal and E is the rationalelectrical potential with a reference at the point of zero charge (PZC).− dγ= σMdE+∑dμ (2.18)Several equations can be derived from eq 2.18. One important relationship that follows from theelectrocapillary equation is: =−∂γ∂μE,μj 6=,T,P(2.19)Eq 2.19 can be used to determine the surface excess (e.g., the extent of adsorption at the interphase)of any species, while keeping the potential and the chemical potential of all other species constant.This equation indicates that measurement of changes in the surface tension as a function of chemicalpotential can yield Gibbs surface excess. Recall that surface excess can also be related to surfacecoverage as shown in eq 2.5 (see Section 2.3.2). However, experimentally it is not trivial to measuresurface/interfacial tension on solid electrodes, so this information must be obtained indirectly from othermeasurements. As a result, we need to consider other relationships.For a solution with constant composition (and at constant temperature and pressure), the partialderivative of γ with respect to potential E gives the excess surface charge density of the metal:σM = −∂γ∂Eμ,T,P(2.20)Eq 2.20 reveals that surface tension is a maximum at the point of zero charge (PZC), which is alsocalled the electrocapillary maximum. At potentials more positive and more negative from the PZC, thesurface tension decreases in a parabolic dependence fashion.163Subsequently, by measuring the change in charge density resulting from a small change in potential,one canmeasure differential capacitance. One definition is shown in eq 2.21, where Cd is the differentialcapacitance. In other words, the ability of the interface to store charge in response to a perturbation inpotential can be characterized by capacitance.Cd =‚∂σM∂EŒ,μ,T,P= −‚∂2γ∂E2Œμ,T,P(2.21)Eq 2.21 relates the surface tension, or excess surface Gibbs energy, to the charge density σM and382.4. Electrochemistrythe differential capacitance Cd. These relationships are purely thermodynamic and are not based on amodel. The only assumption made in these derivations is that the interface is ideally polarizable (i.e.,charge cannot cross the interface). In the next section, the use of capacitance measurements to obtainan estimate of surface coverage will be discussed.2.4.2.2 The Parallel-Plate Model of FrumkinLet us re-consider the Gouy-Chapman-Stern model of the double layer, as represented by eq 2.17 inSection 2.4.1.3. Assuming that measurements are taken in a concentrated solution of supporting elec-trolyte, the contribution from the diffuse layer capacitance can be neglected. Therefore, the double layercapacitance can be approximated by the capacitance of the Helmholtz/Stern layer. It is important tonote that in this inner layer, the electrode is always solvated, and that the solvent molecules (and someions) are in direct contact with the electrode surface. Adsorption onto a site on this surface requires thedisplacement of these solvent molecules and ions. Therefore, electrosorption is a competitive process.For a system that experiences specific adsorption, the inner Helmholtz layer can be modeled as twocapacitors in parallel: (1) the fraction of the surface representing the electrode without any adsorbate(with a capacitance of Cθ=0), and (2) the fraction of the surface representing the electrode completelycovered by the adsorbate (with a capacitance of Cθ=1). The total capacitance is the sum of thesecapacitors in parallel. Therefore, when the electrode is partially covered with the adsorbate, the totalcapacitance of the electrode can be described as a weighted average as shown in eq 2.22.CH = Cθ=0 (1− θ)+Cθ=1θ (2.22)Solving for θ, we get eq 2.23.θ=CH−Cθ=0Cθ=1−Cθ=0(2.23)Recall from Section 2.3.2 that the surface coverage θ is defined as the fractional coverage of themolecule (i.e., θ =/m). Evidently, we can calculate θ with measured CH when both Cθ=0 andCθ=1 are known. Experimentally, Cθ=0 is the capacitance of the electrode in the presence of solventmolecules and ions (i.e., without the adsorbate). CH is the capacitance of the electrode with someadsorbate. Finally, Cθ=1 is the capacitance of the electrode completely covered by the adsorbate (i.e.,at maximum coverage).From eq 2.16 (see Section 2.4.1.1), we see that the capacitance measured depends on the dielectric392.4. Electrochemistryconstant ϵ (if we assume d remains constant). As a result, the Frumkin parallel-plate model can alsobe expressed in terms of ϵ. If we assume for Cθ=0, the dielectric constant measured is that of water(ϵter). For Cθ=1, we assume that the dielectric constant measured is that of the adsorbate (ϵds).When the electrode is partially covered with the adsorbate, the measured CH essentially measures theaverage dielectric constant (ϵg), which gives:ϵg = ϵter (1− θ)+ ϵdsθ (2.24)Most importantly, changes in the electrical double layer due to the adsorption of molecules on theelectrode surface can be measured by differential capacitance measurements (since the dielectric con-stant ϵ also changes, assuming d remains constant). Using eq 2.23, an estimate of the surface coveragecan be obtained.Subsequently, the concentration-dependent surface coverage can be fit to either the Langmuir (eq2.12) or the Frumkin adsorption isotherm (eq 2.14) in order to obtain the free energy of adsorption at aspecific electrode potential. This process can be repeated for different electrode potentials. As a result,potential-dependent free energies of adsorption (ΔG◦ds) can be determined.2.4.2.3 Parabolic Dependence of ΔG◦dsFor the simple Frumkin parallel-plate model, the slope of the ΔG◦dscurve with respect to E is lin-ear. As a result, the ΔG◦dsis parabolically dependent on E.169 This conclusion can be arrived atusing an electrical analogue of charging a capacitor. The energy required to charge a capacitor is´ E0C(θ)EdE =12C(θ)E2. The difference in energy between charging a capacitor with and without anadsorbed monolayer is12(Cθ=0−Cθ=1) (E− EPZC)2, where the potential is referenced to the PZC (incases where it can be determined experimentally).2.4.2.4 PseudocapacitanceIn reality, the excess surface charge density of the metal σM depends on both the electrode potential (E)and the surface excess of a particular species (), which is related to the surface coverage θ. Takingthe differential of σM, which is a function of two variables, yields eq 2.25.dσM =‚∂σM∂EŒdE+‚∂σM∂ŒEd (2.25)Dividing eq 2.25 through by dE yields eq 2.26, where Cd is the measured differential capacitance.402.4. ElectrochemistrydσMdE= Cd =‚∂σM∂EŒ+‚∂σM∂ŒEddE(2.26)The first term of this equation (i.e.,∂σM∂E) is the capacitance of the interface from the double layerunder conditions of constant coverage, as derived in eq 2.21. The second term (i.e.,∂σM∂EddE) is theso-called pseudocapacitance.This pseudocapacitance term is negligible whenddEis small, or at low coverages, which occurs at lowadsorbate concentrations. However, this pseudocapacitance term becomes significant as the concen-tration increases (both  andddEincreases). For example, at more negative electrode potentials, wateradsorbs more strongly onto the metal surface. As a result, water molecules will displace the adsorbateon the surface. This change in coverage will give rise to pseudocapacitance features, representing apeak in the capacitance-potential measurement due to the additional component∂σM∂EddE. A similartrend is also observable at more positive potentials for stable systems.The pseudocapacitance term represents the change in capacitance due to the amount of speciesadsorbed as a function of potential (ddE) multiplied by the electrosorption valency∂σM∂E. In this the-sis, electrosorption is defined as specific adsorption because this process typically involves direct con-tact between an adsorbate and the electrode surface, which occurs when species penetrate the innerHelmholtz or Stern layer (see Figure 2.21). Specific adsorption of molecules or ions causes a change inthe dielectric constant ϵ of the inner layer, since the adsorbed water molecules are replaced by ions ormolecules that may have a different dielectric constant (see eq 2.24). The electrosorption valency termshows that any change in  (and therefore a change in ϵ of the inner layer) will effectively change thecharge density of the metal, which results in a pseudocapacitance feature.It is well known that chloride and hydroxide ions adsorb onto Au surfaces at certain potentials andwill contribute to pseudocapacitance features.170,171 In addition to desorption/adsorption processes,pseudocapacitance peaks can also be caused by phase transitions of the adsorbed layer.172 Never-theless, it is important to note that the Frumkin parallel-plate model does not account for the effects ofpseudocapacitance (i.e.,∂σM∂EandddE), which may limit the potentials that can be analyzed.2.4.2.5 ElectrodesorptionAn adsorbed layer on the electrode can only exist on the surface in a limited potential range (i.e., the Eregion where dγ is a minimum), and the capacitance will be small due to the adsorbed layer (e.g., theadsorbed layer has a lower dielectric constant than the water molecules). At potentials outside of this412.5. Integration of Microfluidics & Electrochemistrywindow, solvent molecules with a high dielectric constant (e.g., water) will displace the adsorbate (i.e.,ddE6= 0), giving pseudocapacitance peaks at potentials near this boundary. In this region, some areasof the electrode are still covered with adsorbate molecules, and some areas of the electrode containonly the electrolyte (i.e., solvent-covered electrode). At extreme potentials, the layer may be completelydisplaced by solvent molecules, rendering a completely solvent-covered electrode. This phenomenonis called electrodesorption.2.5 Integration of Microfluidics & ElectrochemistryIn addition to studying non-faradaic processes, electrochemistry can also be applied to study hetero-geneous electron transfer kinetics173 (typically at the ionic liquid/metal interface).174 Electrochemistryhas been applied in many fields, including corrosion science,175 metallurgy,176 fuel cells,177 semicon-ductors,178 self-assembly of coatings,179 and electrochemical sensors.180 In particular, there is a grow-ing interest in combining microfluidics with electrochemical sensing due to two critical advantages: (1)suitability for miniaturization, and (2) inexpensive experimental setup.108 For example, electrochemicalsystems often use inexpensive potentiostats to apply electric potentials for measuring currents. Fur-thermore, the advances in screen printing (i.e., to increase the manufacturing throughput and decreasethe cost of electrodes)181,182 have aided the development of point-of-care glucose183 and alcohol184sensors.An electrochemical cell typically consists of three electrodes: (1) a working electrode (WE), (2) acounter electrode (CE), and (3) a reference electrode (RE). In most cases, the WE and CE are directlyimmersed in the working solution and the RE is indirectly connected by a conductive salt bridge. Forsensing applications, the electrode size, geometry, material, and surface structure are important factorsto be considered.108,185Most commonly, Au electrodes are patterned on a flat glass substrate for creating microfluidic elec-trochemical systems. Subsequently, a polymer-based structure (i.e., PDMS) with a microfluidic cellchamber is placed in contact with the hybrid glass/gold substrate, forming an enclosed cell chamber(see Figure 1.2 in Section 1.1).186 The design of microfluidic electrochemical systems can have theREs connected via a salt bridge on chip,187 or have the REs and CEs connected externally.188A major problem associated with this type of design is the poor adhesion between gold and PDMS.It is simple to create a strong bond between glass and PDMS as described in Section 2.1.2. However,without treating the gold electrode, the electrolyte may leak at the PDMS−Au interface.57 This leakage422.6. Summaryissue will have numerous adverse effects on the electrochemical measurements.2.5.1 Problems with Electrolyte LeaksSome of the problems associated with electrolyte leaks at the PDMS−Au interface may include:1. Electrolyte leaks will change the electrode surface area under measurement, which will in turnaffect the current measured. For example, when the surface area of the electrode remains un-changed, the increased activity (or concentration) of an electroactive species will result in an in-crease of the current measured. However, an increase in the electrode surface area will also causean increase in the current measured. Therefore, it would be difficult to determine the real cause ofincreased current when both of these factors are present during an electrochemical measurement.2. Impedance is a standard measurement method used in electrochemical biosensors. This methodrelies on the modeling of an electrochemical cell with equivalent circuit elements like resistors orcapacitors. For a system with a leak of electrolyte, the surface area increases (increasing thecapacitance). More importantly, the small crack introduces another resistance that results in avariety of potentials on the electrode surface during current flow due to voltage drops.3. Electrolyte leaks may also result in cross-contamination issues. For instance, solutions with highlydifferent pH values sometimes need to be sequentially introduced and studied. However, anyresidual solution that remain in the unbonded areas at the PDMS−Au interface (e.g., cracks) maysignificantly change the pH of the working solution in the PDMS cell chamber. As a result, theelectrochemical measurements may be adversely affected.2.6 SummaryIn this Chapter, the relevant literature and background theory related to biosensors, surface science,microfluidics and electrochemistry have been reviewed. Next, various surface analysis techniques willbe applied to obtain physicochemical information about TMS-EDTA adsorption on Au. Electrochemicaldifferential capacitance measurements will be used to determine the free energies of adsorption undera complex aqueous environment. Finally, the knowledge acquired will be applied to construct leak-freePDMS-based electrochemical cells.43Chapter 3Surface Analysis of TMS-EDTAAdsorption on Gold3.1 Synopsis2,3This chapter reports on the characterization of TMS-EDTA adsorption on Au via the use of four surfaceanalysis techniques: (1) water contact angle to quantify the wettability of chemically modified Au, (2)X-ray photoelectron spectroscopy (XPS) to confirm the carboxylic acid attachment, (3) atomic force mi-croscopy (AFM) to reveal the surface coverage, uniformity, roughness, and thickness, and (4) infrared(IR) spectroscopy to elucidate the chemical structure of surface-adsorbed species and the extent (i.e.,presence or absence) of siloxane cross-linking. Related species (i.e., 11-MUA, EDTA, and 3-APTMS)were also analyzed to help with the data analysis. Finally, TMS-EDTA-modified Au was applied to de-velop a biosensor surface. Surface plasmon resonance (SPR) was used to test the amount and stabilityof immobilized streptavidin on TMS-EDTA-modified Au following carbodiimide activation.3.2 Surface Analysis3.2.1 Experimental Section3.2.1.1 General Materials and Gold Surface PreparationDetailed lists of the specific materials and reagents used are given below in each of the correspondingexperimental sections. Here, the common materials and surface preparatory procedures are described.The chemical names and structures of the molecules used in this chapter are presented in Figure 3.1. Allchemicals and reagents were obtained from commercial sources: N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acid, Na salt (TMS-EDTA; 50% in water) was from United Chemical Technologies (Bris-tol, PA); 11-mercaptoundecanoic acid (11-MUA; 95%), (3-aminopropyl)-trimethoxysilane (3-APTMS;443.2. Surface AnalysisFigure 3.1: Chemical structures of the molecules used in this chapter: TMS-EDTA, 11-MUA, EDTA, and3-APTMS.97%), ethylenediaminetetraacetic acid (EDTA; BioUltra, ≥99.0%), acetone (ACS reagent, ≥99.5%) and2-propanol (ACS reagent, ≥99.5%) were from Sigma-Aldrich (Oakville, ON, Canada). Ethyl alcohol (de-natured) was purchased from UBC Zoology Stores (Vancouver, BC). Ultrapure water (∼18.2 MΩ-cm)produced by a Milli-Q water purification system (EMD Millipore) was used.The planar Au slides (see additional details in the corresponding experimental sections below) werecleaned by immersion in acetone, 2-propanol, and ultrapure water (repeated three times), followed byblow drying in an argon or nitrogen stream (Ultra High Purity 5.0, Praxair Canada Inc.). Subsequently,these planar Au surfaces were immediately immersed into different reagent solutions (specified below)for surface chemical modifications.3.2.1.2 Water Contact Angle and X-ray Photoelectron Spectroscopy (XPS)For these measurements, Au substrates were manufactured in-house. The glass slides were cleaned ina piranha solution (5:1 H2SO4:H2O2) at 80 °C for 15 min, rinsed with DI water, and dried in a N2stream.These slides were subsequently placed into a 200 °C oven for 15 min to eliminate residual moisture.A 20 nm chromium adhesion layer and a 200 nm gold layer were sequentially evaporated onto the453.2. Surface Analysisclean glass slides in an electron-beam evaporator. After cleaning with organic solvents as describedabove, three Au slides were separately immersed for 2 hours in (1) aqueous solution of 10% TMS-EDTA(v/v), (2) ethanolic solution of 10 mM 11-MUA, and (3) ultrapure water (serving as reference surface).The surface-modified, gold-coated glass substrates were then rinsed with the respective solvent, driedunder a N2stream, and analyzed immediately. The contact angle analysis was performed using theLow-Bond Axisymmetric Drop Shape Analysis (LB-ADSA) Plugin for ImageJ.189 The peaks in the XPSspectra were individually fitted assuming a Gaussian distribution.3.2.1.3 Atomic Force Microscopy (AFM)For AFM studies, cleaned planar Au slides (TA134, 5 nm Ti and 100 nmAu) from Evaporated Metal Films(Ithaca, NY) were used. An Agilent 5500 AFM equipped with a 90 μm scanner and SiN Nanoprobe tips(Digital Instruments, spring constant of 0.8 N/m nominal) were used for these measurements.AFM Contact Mode Imaging For contact mode imaging, two types of surfaces were created. Thefirst type was prepared by immersing the Au slide in an aqueous solution of 10% TMS-EDTA (v/v), andthe second type was prepared by immersing the Au slide in ultrapure water (serving as Au substrate).After 2 hours of immersion, the surfaces were rinsed with ultrapure water, blown dry with argon gas, andanalyzed immediately. A 512×512 pixel grid was measured over an area of 3×3 μm2 at 1 Hz per linewith integral and proportional gains of 0.5 and 1, respectively (with a setpoint in the attractive regime).Topography and lateral force measurements were obtained for both TMS-EDTA-coated Au and clean Ausubstrate. Data processing was performed in Gwyddion 2.3.4 (Czech Metrology Institute) with medianheight matching and second-order polynomial baseline subtraction.AFM Force-Distance Spectroscopy For force-distance spectroscopy, a third type of surface wasprepared (in addition to the two surfaces used for contact-mode imaging) by immersing a cleaned Auslide in an ethanolic solution of 10 mM 11-MUA for 2 hours. For the three types of freshly preparedsurfaces, approach and retraction force curves were obtained, and sampling was conducted at 64 pointsdistributed in an 8×8 grid over the same scan size as contact mode imaging. The tip approach speedwas 117 nm/s to allow for adequately high resolution sampling at 12.5 kHz. Data processing for theforce curve measurements was conducted in MATLAB. Briefly, the 64 approach-only force curves wereextracted, and the first derivative of the tip deflection (V) to distance (nm) was obtained. Subsequently,the maximum of these values was determined and the corresponding distance value was defined as the463.2. Surface Analysisinterface of the gold (i.e., distance = 0 nm). The force-distance curves from these three samples werecompared.3.2.1.4 Infrared SpectroscopyPM-IRRAS For polarization modulation-infrared reflection-absorption spectroscopy (PM-IRRAS) stud-ies, planar Au slides (TA134, 5 nm Ti and 100 nm Au) from Evaporated Metal Films (Ithaca, NY) werecleaned using the protocol described above. Three types of Au substrates were created by overnightimmersion in (1) aqueous solution of 10% TMS-EDTA (v/v), (2) aqueous solution of 10% 3-APTMS (v/v),and (3) ethanolic solution of 10 mM 11-MUA. The Au slides were blown dry with argon gas, and storedin a desiccator to remove excess moisture for at least two days prior to the experiments.PM-IRRAS was performed using a Bruker-55 spectrometer with an external PMA 50 accessory.The IR beam, after passing through a ZnSe grid polarizer and a ZnSe photoelastic modulator (HINDSInstruments, PEM-90, modulation frequency of 50 kHz), was focused on the sample at an incident angleof 80−85°. The light reflected from the sample was then focused onto an MCT detector (model D313/x1,Infrared Associates Inc. Stuart, Fl, USA.). The PM-IRRAS signal is given by the differential reflectivityΔRR=(Rp−Rs)(Rp+Rs)and the presented spectra resulted from the sum of 2048 scans (from at least two co-added spectra) recorded with 4 cm−1 resolution. The final step of data processing involved baselinecorrection using spline interpolation of the experimental data points in MATLAB.ATR-FTIR Three chemical samples (3-APTMS, EDTA, and TMS-EDTA) were analyzed by attenuatedtotal reflectance Fourier transform infrared (ATR-FTIR). The spectra of the samples were recorded usinga Perkin-Elmer Frontier Spectrometer (4 scans and 4 cm−1 resolution, model equipped with a UniversalATR Sampling Accessory). The 3-APTMS and TMS-EDTA samples were separately prepared by pipet-ting ∼1 mL of bulk solution onto glass slides containing KBr powder. The third sample was preparedby mixing solid EDTA with KBr powder on a new glass slide. The three powdered mixtures were sealedinside a desiccator with desiccants for at least two days before analysis in order to remove excessivewater.473.2. Surface Analysis3.2.2 Results and Discussion3.2.2.1 Water Contact Angle and X-ray Photoelectron Spectroscopy (XPS)Water contact angle and X-ray photoelectron spectroscopy (XPS) were used to examine the feasibilityof using TMS-EDTA to chemically modify Au substrates for creating a carboxyl-terminated Au surface.This surface was compared with 11-MUA-modified Au. 11-MUA is an extensively studied alkanethiolthat has a well-defined orientation on Au (i.e., with terminating carboxylic acid groups).96Water contact angle measurements using the sessile drop method were performed on the surfacesof bare gold (Figure 3.2A), 11-MUA-modified gold (Figure 3.2B,) and TMS-EDTA-modified gold (Figure3.2C). For proof-of-concept, a single measurement was performed on each sample. The Au surfacesmodified with TMS-EDTA and 11-MUA had water contact angle values that were more hydrophilic thanthe clean Au, confirming chemical modification (see Section 2.2.1).XPS was used to further characterize the chemically modified Au surfaces. The results are shownin Figure 3.3. Figure 3.3A reveals that the gold surface modified with 11-MUA shows an increase in thequantity of carbon (C) in comparison to the bare gold control surface. The presence of some C atomson the gold control surface may be from the organic solvents used or from the organics in air. The highresolution spectrum shows a C 1s peak at 284.7 eV with a shoulder peak at 288 eV, corresponding tothe terminating carboxyl groups of 11-MUA (see Section 2.2.5.2).190,191 Similarly, Figure 3.3B shows anincrease in the quantity of C for TMS-EDTA-modified gold substrate and a shoulder that corresponds tocarboxyl groups (as compared to the bare gold surface). These results suggest that the carboxyl groupsare still present on the TMS-EDTA-modified Au surface.3.2.2.2 Atomic Force Microscopy (AFM)Water contact angle and XPS results suggest that TMS-EDTA can be used to chemically modify Au andcreate terminating carboxyl groups. Characterization of the surface topography is needed to better un-derstand the surface coverage and uniformity of TMS-EDTA adsorption on gold. Spatial characteristicssuch as film uniformity and film thickness are particularly important for biosensing and biosensor fabri-cation. For example, the uncontrolled deposition of silanes onto silicon and glass surfaces can result infilms with a loose network structure, and the final morphology and thickness (typically 1 to 8 molecularlayers) are sensitive to water content, pH, and temperature.60,192,193 To explore the thickness and uni-formity of adsorbed TMS-EDTA layer(s) on Au, atomic force microscopy (AFM) was used (see Section2.2.2).483.2. Surface AnalysisFigure 3.2: Water contact angle measurements on (A) bare gold, (B) 11-MUA-modified gold, and (C)TMS-EDTA-modified gold.493.2. Surface AnalysisFigure 3.3: XPS high resolution spectra of the C 1s signal for (A) 11-MUA and (B) TMS-EDTA on gold(with spectrum for bare gold surface).503.2. Surface AnalysisFigure 3.4: Topography (A) and lateral friction (B) maps from AFM contact mode imaging are shown(scanned using the same tip for all images). Bare Au substrate (top row); TMS-EDTA-modified Au(middle row); and the bare Au substrate after scanning the TMS-EDTA layer (bottom row).AFMContact Mode Imaging Topography (Figure 3.4A) and lateral force or friction (Figure 3.4B) mea-surements obtained from AFM contact mode imaging of a clean Au substrate and a TMS-EDTA-modifiedAu are shown. Histograms calculated from these topography and friction measurements were also de-termined (see Figures 3.5A and 3.5B). From Figures 3.4A and 3.5A, it is evident that the topography ofthe Au surface appears smoother when coated with TMS-EDTA. Similarly, from Figures 3.4B and 3.5B,the Au surface exhibits a higher lateral friction when coated with TMS-EDTA.Subsequently, the same tip was used to re-scan the clean Au substrate (after scanning the TMS-EDTA-modified Au) to ensure comparability of the results. It is interesting to note that the friction of thebare Au substrate (after TMS-EDTA sample scan) has increased slightly (as compared with the first bareAu substrate scan). This result indicates that the silicon nitride tip may have been contaminated withadsorbed TMS-EDTA molecules. Nevertheless, it is evident from Figures 3.4 and 3.5 (as well as scansfrommultiple areas on the surface) that TMS-EDTA coats Au uniformly as measured over a large surfacearea (scan size of 3×3 μm2). Moreover, TMS-EDTA does not form large clumps on the Au surface (i.e.,no patchy areas were observed) as commonly observed on silicon-based surfaces.194AFM Force-Distance Curve Next, AFM force-distance curve was used to provide a rough estimateof the relative thickness of the TMS-EDTA-coated surface, as compared to a control sample of 11-MUA-coated Au. In these measurements, the cantilever and the tip were moved directly toward the513.2. Surface AnalysisFigure 3.5: Topography (A), and lateral friction (B) distributions calculated from AFM contact modeimages (scanned using the same tip for all images): cleaned gold substrate before sample scan (red),TMS-EDTA modified gold substrate scan (black), and cleaned gold substrate after sample scan (blue).523.2. Surface Analysissample until a contact is made, and then retracted. The interaction between the tip and the sample wascontinuously monitored (see Section 2.2.2.2). To illustrate the procedure used for data analysis, the 64approach-only AFM force curves (cantilever deflection vs piezo movement) for TMS-EDTA-modified Ausubstrate are shown in Figure 3.6A. First derivatives of the force curves are shown in Figure 3.6B.Subsequently, the maximum of these derivatives was determined and the corresponding distancevalue was defined as the interface of the gold (i.e., distance = 0 nm) as shown in Figure 3.7A. Thered curve shows the averaged values for the 64 samples. Figure 3.7B shows a zoomed-in view of theaveraged and normalized values, with the blue line serving as the reference line. The same procedurewas applied to analyze the force-distance curves of 11-MUA-modified Au and bare Au substrate.Figure 3.8 shows the zoomed-in view of the interactions of silicon nitride tips with TMS-EDTA-coatedgold (red), 11-MUA-coated gold (blue), and cleaned gold substrate (black). It is observed that the curvefor 11-MUA-coated Au remained constant at a distance from 3 nm to ∼1.6 nm, and started to increasefrom∼1.6 nm until the interface of the gold was reached (distance = 0 nm). This result confirmed that the11-MUA on Au surface was a monolayer195 and validated this approach for determining the thickness ofadsorbed layer of TMS-EDTA. From Figure 3.8, the force curves for the two chemically modified surfacesare indistinguishable (but distinctly different from the clean Au surface), suggesting that the thickness ofTMS-EDTA on Au is similar to the 11-MUA monolayer.The AFM results reveal that TMS-EDTA uniformly coats the planar Au surface and has a thicknesscomparable to a 11-MUA monolayer on Au, suggesting that TMS-EDTA adsorption is not multilayer innature. Next, infrared spectroscopy was utilized to better understand the chemical nature of TMS-EDTAadsorption on Au.3.2.2.3 Infrared SpectroscopyWater contact angle and XPS results provided an initial analysis of the chemical nature of TMS-EDTAadsorption on Au, and suggested the presence of terminating carboxyl groups. However, the stabilityof the adsorbed TMS-EDTA layer in ultra-high vacuum (UHV), as required by the XPS measurements,might not be sufficient for quantitative analysis because of exposure to X-rays or low pressure. It is impor-tant to note that the solution-based chemical modification with TMS-EDTA should always be performedat room temperature, under atmospheric pressure. Furthermore, the TMS-EDTA-coated gold surfacewould not be exposed to UHV in future biosensor fabrication and applications. Therefore, TMS-EDTAadsorption on Au needs to be analyzed under ambient conditions in order to obtain a more accuratepicture of the chemical structure of the adsorbed layer.533.2. Surface AnalysisFigure 3.6: (A) An example of the 64 approach-only AFM force-distance curves for TMS-EDTA-modifiedAu. (B) First derivatives of the force-distance curves.543.2. Surface AnalysisFigure 3.7: (A) The maximum of the derivatives was defined as the surface of Au (i.e., distance = 0 nm).The red curve shows the averaged values for the 64 samples. (B) A zoomed-in view of the averagedand normalized values, with the blue line serving as the reference line.553.2. Surface AnalysisFigure 3.8: The zoomed-in view of silicon nitride tip interaction with TMS-EDTA-coated gold (red), 11-MUA-coated gold (blue), and clean gold substrate (black).PM-IRRAS Polarization modulation-infrared reflection-absorption spectroscopy (PM-IRRAS) was usedto produce a spatially averaged picture of the chemical structure of the adsorbed layer on Au, underambient conditions. All PM-IRRAS spectra presented have been baseline-corrected in MATLAB. Anexample of the baseline correction method applied for TMS-EDTA spectrum is shown in Figure 3.9. InFigure 3.9A, the raw PM-IRRAS signal (solid line) is plotted with the baseline curve determined fromspline fitting (dotted line). The asterisks were the manually selected points for creating the baselinecurve. In Figure 3.9B, the baseline-subtracted signal is shown. Subsequently, this signal was correctedfor sample gain(s) to obtain the absorbance spectrum.The amount of TMS-EDTA adsorbed and the chemical structure of the adsorbed layer (e.g., surfacereactions that might change the chemical nature of the layer) could be inferred by comparison withother self-assembled monolayer (SAM) systems.196 In particular, 11-MUA SAMs have been studiedusing PM-IRRAS.197,198 This information supports the ability of PM-IRRAS to identify carboxyls andalkyl groups on the Au surface. In addition, the 3-APTMS-modified Au was also analyzed to providesome information regarding siloxane cross-linking on Au.152 Relative strengths of the absorbance willalso provide a rough estimate of the extent of TMS-EDTA adsorbed on the surface, keeping in mind thatthe PM-IRRAS signal strength depends on the molecular orientation (see Section 2.2.3). The chemicalstructures of 3-APTMS, 11-MUA, and TMS-EDTA are shown in Figure 3.1. The PM-IRRAS spectra forAu modified with these substances are shown in Figure 3.10 for the alkyl stretching region (3200–2700563.2. Surface AnalysisFigure 3.9: An example of background correction method used for PM-IRRAS data analysis. (A) Theraw signal of TMS-EDTA-modified Au (solid line) is shown with background curve determined from splinefitting (dotted line). (B) The background-subtracted signal is shown. Subsequently, this signal wascorrected for sample gain(s) to obtain Absorbance spectrum.573.2. Surface AnalysisFigure 3.10: The PM-IRRAS spectra for 3-APTMS on Au (black), 11-MUA on Au (red), and TMS-EDTAon Au (blue). Alkyl stretching region (3200–2700 cm−1) and the fingerprint region (1800–800 cm−1)are shown.cm−1) and the fingerprint region (1800–800 cm−1).ATR-FTIR Attenuated total reflectance Fourier transform infrared (ATR-FTIR) was used to provide sup-porting information regarding the chemical structure and the corresponding infrared (IR) bands of thebulk chemical compounds (see Section 2.2.3.1). In the literature, solid 11-MUA has been previously an-alyzed by ATR-FTIR and these results are used without confirmation in this study.199–201 The chemicalsEDTA and 3-APTMS (see Figure 3.1) were analyzed to assist in identifying the IR bands of carboxylatesand silanes, respectively. The ATR-FTIR spectra of 3-APTMS, EDTA, and TMS-EDTA were measuredand presented in Figure 3.11. Alkyl stretching region (3200–2700 cm−1) and the fingerprint region(1800–800 cm−1) are shown.PM-IRRAS and ATR-FTIR Analysis First, TMS-EDTA and 11-MUA spectra will be compared andcontrasted, which will then be followed by a similar analysis of TMS-EDTA and 3-APTMS spectra. Con-sidering the 11-MUA spectra in Figure 3.10, this SAM shows an absorption that is characteristic of CH2symmetric and asymmetric stretching bands (2850 and 2928 cm−1, respectively) as expected due toits long hydrocarbon chain.198 The absorption band around 1700 cm−1 is from the C−O stretch for11-MUA on Au, but may also include adsorbed water, which may still be present on the surface evenafter extensive drying in the desiccator. The small peak at 1470 cm−1 was assigned to the CH2de-formation.198 Considering the TMS-EDTA spectra in Figure 3.10, the CH2symmetric and asymmetricstretching bands (also evident from ATR-FTIR spectra as shown in Figure 3.11A) are also present as583.2. Surface AnalysisFigure 3.11: Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) transmis-sion spectra of bulk TMS-EDTA (blue), EDTA (green), and 3-APTMS (red) mixed with KBr powder. (A)Alkyl stretching region (3200–2700 cm−1) and (B) the fingerprint region (1800–800 cm−1) are shown.593.2. Surface Analysisexpected (2850 and 2928 cm−1, respectively), but a smaller absorbance than 11-MUA is observed.Signatures of CH3asymmetric stretching mode198 are also evident as a shoulder at 2960 cm−1, sup-porting the conclusion that the methyl groups may still be present in the adsorbed layer (a small 2960cm−1 shoulder is also visible in the 11-MUA spectrum, possibly due to the solvent used). The presenceof the absorption at 1680 cm−1 is evidence that the carboxyl groups are still present in the adsorbedTMS-EDTA. These carboxyl groups are also present in the ATR-FTIR spectra of EDTA and TMS-EDTA,and are absent in the 3-APTMS spectra (as shown in Figure 3.11B). Furthermore, the 11-MUA SAM isknown to form a thin monolayer on Au surface. The signal intensities of TMS-EDTA and 11-MUA spectrain Figure 3.10 are comparable (i.e., both have been multiplied by a factor of 4). This result suggests thatthe thickness of the TMS-EDTA layer is similar to that of the 11-MUA monolayer.The PM-IRRAS spectra of 3-APTMS (Figure 3.10) have distinct features in the 1000–1200 cm−1region (also evident from ATR-FTIR spectra as shown in Figure 3.11B), which are characteristic ofSi−O−Si asymmetric stretching as well as Si−O−C bands196 due to polysiloxane formation (see Sec-tion 2.1.1.2). The extent of siloxane cross-linking for 3-APTMS is quite substantial, as shown by thelarge peak at 1130 cm−1 and the lack of CH3features in the alkyl region (see Figure 3.10).152 This isevidence of hydrolysis of the methoxy groups and siloxane cross-linking (creating a significant amount ofSi−O−Si absorption). These results are consistent with the previous observation that water plays an im-portant role in the surface attachment of amino-silanes. More specifically, previous results suggest thatin the presence of water the final multilayer film of amino-silanes on Au is composed of a cross-linked,two-dimensional siloxane network (due to complete hydrolysis).152 On the other hand, the presence ofsome polysiloxane formation in TMS-EDTA (also evident from Figure 3.11B) is not too surprising dueto high pH values (see Appendix A), but the extent is significantly less than the 3-APTMS-coated Ausample (see Section 2.1.1.2). Furthermore, the intensities of the bands of 3-APTMS spectra are morethan 5X that of TMS-EDTA. These results confirm the multilayer formation of 3-APTMS on Au,152,192and suggest that the amines are adsorbed onto the Au surface and the silane groups are directed away,resulting in extensive cross-linking.202,203In these IR studies, the TMS-EDTA-coated gold surface shows a significant presence of carboxylgroups (similar to 11-MUA coated Au) and a lack of polysiloxane formation (in contrast to 3-APTMScoated Au). However, the orientation of adsorbed TMS-EDTA on Au remains unclear. Subsequently,surface plasmon resonance (SPR) was used to determine the chemical functionality of TMS-EDTA-modified Au, and thereby provide at least a rough indication of orientation.603.3. Surface Plasmon Resonance (SPR)3.3 Surface Plasmon Resonance (SPR)The adsorption of TMS-EDTA on Au has been characterized via the use of four surface analysis tech-niques. Water contact angle, XPS, and IR results show that the carboxyl groups are present on TMS-EDTA-modified Au, and AFM results show that this surface is uniform. However, the orientation ofadsorbed TMS-EDTA on Au still remains unclear. The orientation of the carboxyl groups is importantfor sensor fabrication and biosensor applications. In particular, if the carboxyl groups are oriented awayfrom the gold surface, then they can potentially be used to covalently couple substances with primaryamines, following carbodiimide activation. Therefore, SPR experiments (see Section 2.1.3) were con-ducted to demonstrate the feasibility of using TMS-EDTA-modified Au for coupling biomolecules andto determine its relative stability (compared to bare Au) when a stringent wash buffer was used. Thequantity of biomolecules captured and the stability of this layer are important factors to be consideredwhen developing a robust optical biosensor. Streptavidin was used as a model protein because it’s acommon component of chemically selective films for biosensing applications.553.3.1 Experimental SectionFor SPR studies, two planar Au sensor chips (SIA Kit Au) from GE Healthcare (Mississauga, ON,Canada) were cleaned as previously described. The Au chip was either immersed in (1) aqueoussolution of 10% TMS-EDTA (v/v), or (2) ultrapure water (serving as a reference surface for physicalimmobilization). After 2 hours of immersion, the surfaces were rinsed with water and blown dry withargon gas and then glued onto the cassettes according to the manufacturer’s instructions. All SPRexperiments were performed at 25 °C in an SPR biosensor (BIACORE 3000 apparatus) using these in-house assembled sensor chips. 1X PBS, pH 7.4, was prepared using Dulbecco’s phosphate-bufferedsaline (powder, no calcium, no magnesium) from Life Technologies (Burlington, ON, Canada) and usedas the running buffer. N-(3-(Dimethylamino)propyl)-N´-ethylcarbodiimide hydrochloride (EDC; 98%) andN-hydroxysuccinimide (NHS; 98%) were from Sigma-Aldrich (Oakville, ON, Canada). Streptavidin pu-rified (S203) from Leinco Technologies, Inc. (St. Louis, MO) was used as the model protein. For theAu chip immersed in pure water, streptavidin (100 μg/mL, diluted in PBS) was injected over the surfacefor 10 min to facilitate physisorption. For the TMS-EDTA-modified Au chip, carboxyl groups were firstactivated with an aqueous mixture of 50 mM NHS and 200 mM EDC for 7 min, and streptavidin wasinjected over the surface for 10 min (see Section 2.1.3.1) to facilitate chemisorption. The stability ofboth streptavidin-covered surfaces was investigated by introducing five short injections (3 min each) of613.4. Conclusionsa stringent regeneration buffer (50 mM NaOH). The NaOH regeneration buffer (NaOH 50) was from GEHealthcare. All buffer and sample injections were conducted at a flow rate of 10 μL/min.3.3.2 Results and DiscussionFigure 3.12A shows the SPR sensorgram that indicates the immobilization of streptavidin on a bare Auchip, followed by five injections of 50 mM NaOH regeneration buffer. Figure 3.12B shows the immobi-lization of streptavidin on a TMS-EDTA-modified Au chip (using NHS/EDC coupling chemistry), followedby five injections of the same regeneration buffer.As can be seen from Figure 3.12A, streptavidin immobilized on the bare gold surface resulted in apeak value of about 690 RU (Resonance Unit or Response Unit), and streptavidin washed off of the Ausurface after NaOH injection resulted in a decrease of about 630 RU. Therefore, only about 9.1% of thesignal for adsorbed streptavidin remained after five NaOH injections. In contrast, streptavidin immobi-lized on a TMS-EDTA-modified gold surface (Figure 3.12B) resulted in a peak value of about 1610 RU,and streptavidin washed off after NaOH injections resulted in a decrease of about 600 RU. Using a simi-lar calculation, about 62.7% of adsorbed streptavidin remained on the TMS-EDTA-modified Au after thestringent NaOHwash. Thus, the TMS-EDTA-modified gold surface yielded both a substantial increase inthe amount of streptavidin immobilized and a substantial decrease in the amount of streptavidin removedafter NaOH injections. The results reported here suggest that TMS-EDTA-modified Au consists of freecarboxyl groups that are able to react with primary amino groups. Previous publications demonstratingsilanes interacting with gold surfaces have been reported, supporting this proposed orientation.149,2043.4 ConclusionsThe adsorption of a carboxylated silane (TMS-EDTA) on Au has been characterized using water contactangle, XPS, AFM, and IR methods. Water contact angle and XPS results provide an initial analysis ofthe chemical nature of TMS-EDTA adsorption on Au, and suggest the presence of carboxyl groups. AFMimaging strongly suggests that a thin and uniform coverage of TMS-EDTA on the gold surface is obtained.In IR studies, the TMS-EDTA-coated gold surface shows a significant presence of carboxyl groups (sim-ilar to 11-MUA-coated Au) and a lack of polysiloxane formation (in contrast to 3-APTMS-coated Au).Finally, the results from SPR experiments indicate that free carboxyl groups on TMS-EDTA-modified Auare available for the immobilization of streptavidin, after activation by NHS/EDC. Furthermore, this un-conventional surface chemistry can withstand stringent regeneration conditions—a quality important for623.4. ConclusionsFigure 3.12: (A) SPR sensorgram showing the immobilization of streptavidin on a bare Au chip, followedby five injections of the 50mMNaOH regeneration buffer. (B) Streptavidin immobilization on TMS-EDTA-modified Au chip using NHS/EDC chemistry, followed by five injections of the same regeneration buffer.633.4. Conclusionsdeveloping robust biosensors. These results suggest that at least some of the carboxyl groups are ori-ented away from the surface. This orientation serves to facilitate the coupling reactions that are neededfor the immobilization of biomolecules via NHS/EDC chemistry. The data presented so far show thatTMS-EDTA adsorption on Au can be used to create a useful biosensing surface, but an estimate of itsGibbs free energy of adsorption onto Au is needed. Therefore, electrochemical differential capacitancewas used to obtain this information as presented in the following chapter.64Chapter 4Thermodynamic Studies of TMS-EDTAAdsorption on Au usingElectrochemical Methods4.1 Synopsis3The Gibbs free energies of TMS-EDTA adsorption on Au (in the presence of water and ions) are quan-tified in this chapter due to their importance to PDMS bonding and biosensing applications. Electro-chemical differential capacitance is one of the best methods to determine the standard free energies ofadsorption (see Section 2.3) in a complex aqueous environment, while also elucidating its dependenceon the substrate potential (see Section 2.4). Therefore, in this chapter, a determination of the free energyof adsorption of TMS-EDTA onto Au (in an aqueous electrolyte similar to the deposition solution usedpreviously) using electrochemical differential capacitance measurements is described.4.2 Experimental Section4.2.1 General Materials and Gold Surface PreparationAll chemicals and reagents were obtained from commercial sources: N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acid, Na salt (TMS-EDTA, pH ∼11; 50–60 wt % water, 32–38 wt % TMS-EDTA salt, and8–12 wt % sodium chloride) was from United Chemical Technologies (Bristol, PA). Sodium phosphatedibasic dihydrate (puriss. p.a., buffer substance), potassium chloride (puriss. p.a. ACS), and potassiumhydroxide (pellets, 99.99% metals basis, semiconductor grade) were from Sigma-Aldrich (Oakville, ON,Canada). Ultrapure water (∼18.2 MΩ-cm) produced by a Milli-Q water purification system (EMD Mil-lipore) was used. Polycrystalline Au bead electrodes (diameter ∼=2 mm) were cleaned and flame654.2. Experimental Sectionannealed by heating in butane flame until red hot, followed by rinsing with ultrapure water (repeatedthree times).4.2.2 Electrochemical Measurements4.2.2.1 InstrumentationElectrochemical experiments were performed in a three-electrode glass cell with an Autolab electro-chemical analyzer (PGSTAT 12, Eco Chemie B.V.) using NOVA 1.8 software. A scan/sweep controller(EG&G PAR 175) and an analog lock-in amplifier (EG&G model 5210) were used for cyclic voltammetry(CV) and differential capacitance experiments. The working electrode was a Au bead; the counter elec-trode was a Pt coil; and the reference electrode was a saturated calomel electrode (SCE). The referenceelectrode was connected to the electrolyte buffer solution via a salt bridge. The working solution wasdeoxygenated with argon and purged over the surface in all experiments.4.2.2.2 Working SolutionA buffered working solution (an electrolyte containing 100 mM phosphate, 500 mM KCl, and 65 mM KOHat pH 11.6), hereafter called DiffCap buffer, was prepared and used as the electrolyte for all electrochem-ical differential capacitance measurements. This buffer was formulated to simulate the multi-componentsurface modification solution used for chemically modifying the Au substrates. A final working solutionvolume of 50 ± 0.1 mL was used in all experiments (after compensating for volume of solution used insalt bridge).4.2.2.3 Differential Capacitance MeasurementsTo ensure high reproducibility and cleanliness of the three-electrode electrochemical setup, a cyclicvoltammogram (CV) of the gold bead working electrode in DiffCap buffer was obtained before and afterthe addition of bulk TMS-EDTA concentrations for all experiments (see Figures B.1 and B.2 of AppendixB). Incremental concentrations of TMS-EDTA were then added to the DiffCap buffer. At each concentra-tion, capacitance wasmeasured for a range of 19 potentials (starting from−1.1 V and stepping positivelyto 0.2 V). At each potential, stirring was turned on for 60 s and turned off prior to measuring the capaci-tance (repeated six times). After 6 min of stirring, the potential-dependent adsorption of TMS-EDTA onAu was assumed to have reached an equilibrium (see Figures B.3 to B.9 of Appendix B). To desorb theTMS-EDTA layer (true for concentrations less than 50 mM) and to ensure similar starting conditions, a664.3. Results and Discussion−1.1 V potential was applied (for 30 s without stirring) before proceeding from one potential to the next(see Table B.1 of Appendix B). The resulting capacitance was determined assuming a series RC equiv-alent circuit. The capacitance values measured after 6 min of stirring were then averaged and used forfurther data analysis.4.2.2.4 Data AnalysisMATLAB was used to analyze the concentration- and potential-dependent differential capacitance datafor TMS-EDTA adsorption on a Au bead electrode. Using the estimation method based on Shepherdet al.’s study,205 the capacitance measured in DiffCap buffer solution at −0.9 V was normalized tothe area of the Au bead electrode (calculated and estimated to be ∼0.25 cm2), and a value of 16.4μF/cm2 was obtained. All capacitance data were adjusted to account for the Au electrode area to ensurecomparability of the results. Subsequently, differential capacitance values were converted to surfacecoverage (at each applied electrode potential) using the Frumkin parallel plate capacitor model206 asshown in eq 4.1 (see Section 2.4.2.2).Cθ = Cθ=0(1− θ)+Cθ=1θ (4.1)Here, Cθ is the measured differential capacitance value at a particular TMS-EDTA concentration;Cθ=0 is the differential capacitance value of DiffCap buffer (i.e., zero TMS-EDTA coverage); Cθ=1 is theminimum differential capacitance value obtained from 200 mM TMS-EDTA (i.e., maximum TMS-EDTAcoverage); and θ is the calculated fractional surface coverage. Next, the concentration-dependent sur-face coverage at each applied electrode potential was fit to Langmuir or Frumkin adsorption isotherms toobtain potential-dependent free energies of adsorption (−ΔGods) and Frumkin lateral interaction param-eters (α),207 where positive or negative values of α indicate a repulsive or attractive interaction (betweenadsorbed molecules), respectively (see Section 2.3).4.3 Results and Discussion4.3.1 Open-Circuit Potential (OCP) During Surface ModificationFrom Figure 4.1, the presence of dissolved O2was found to set the potential of the substrate during theTMS-EDTA surface modification procedure. Specifically, the open-circuit potential (OCP) was measuredto be ∼0.05 V and ∼ −0.025 V (vs SCE) in the presence and absence of O2, respectively. In all of the674.3. Results and DiscussionFigure 4.1: The open-circuit potential (OCP) of bare Au bead electrode immersed in 10% TMS-EDTA(v/v) solution with (dotted curve) and without (solid curve) oxygen were determined.previous surface analyses (see Chapter 3), the Au substrates were functionalized with an aqueous 10%TMS-EDTA (v/v) solution (prepared using the commercially available stock TMS-EDTA solution—pH∼11—containing 50–60 wt %water, 32–38 wt % TMS-EDTA salt, and 8–12 wt % sodium chloride), in thepresence of dissolved O2. These conditions are important when determining the Gibbs free energies ofTMS-EDTA adsorption under the typical wet chemical coating conditions. In other words, the electrolytebuffer must replicate the conditions of the aqueous 10% TMS-EDTA (v/v) solution in order to determinethe relevant Gibbs free energies of TMS-EDTA adsorption on Au.4.3.2 Initial Measurements of Au Electrochemistry in Working BufferFor an accurate characterization of TMS-EDTA adsorption on Au, the electrolyte composition must beheld constant. However, with every increase in TMS-EDTA concentration, [Cl− ] also increased (themolar ratio of sodium chloride to TMS-EDTA salt of the stock solution was calculated to be ∼2.26).To minimize the effects of chloride competitive adsorption (contributed from the incremental addition ofstock TMS-EDTA solution), 500 mM KCl was added to the 100 mM phosphate buffer. Furthermore, 0.2g of KOH was added to the phosphate and chloride buffer solution to replicate the highly basic stateof 10% TMS-EDTA (v/v) aqueous solution (pH = 10.88 ± 0.03). Consequently, the DiffCap buffer was684.3. Results and Discussionprepared and used in these electrochemical differential capacitance studies. The capacitance of thegold bead electrode in DiffCap buffer was measured at the start of each experiment and demonstratedhigh reproducibility (see four examples shown in Figure 4.2A). Figure 4.2B shows the averaged rawDiffCap Buffer data and the corresponding standard deviations. These standard deviation values wereused to estimate the errors in subsequent fittings.4.3.3 Data Treatment and AnalysisDifferential capacitancemeasurements were performed on gold with incremental concentrations of TMS-EDTA (1 μM to 200 mM) added to the DiffCap buffer. At each concentration, a series of 19 electrodepotentials were applied (changing from −1.1 to 0.2 V), and differential capacitance was measured afterwaiting for 6 min with stirring at each potential, which resulted in an experimental duration of approx-imately 2 hours. Therefore, further tests were conducted to test the stability of TMS-EDTA-coated Auelectrode in DiffCap buffer for several hours. When compared with a bare Au electrode, the capaci-tance of TMS-EDTA-modified electrode appeared to be more stable in DiffCap buffer (due to a minimaldecrease in capacitance) for the duration of the experiments (see Figure 4.3).Triplicate measurements of lower concentrations of TMS-EDTA were conducted (an example isshown in Figure 4.4). Small differences were observed, indicating that the measurements were de-pendent on the area of the Au bead immersed in the buffer solution. Therefore, an area-normalizationprocedure for the Au bead electrode was required.The area of the Au bead electrode was estimated using the method based on Shepherd et al.’sstudy,205 by comparing the capacitance of a Au bead electrode at −0.9 V in both 50 mM PerchlorateBuffer (basic pH) and DiffCap Buffer (see Figure B.10 of Appendix B). An electrode area of ∼0.25 cm2was determined.All of the subsequent capacitance data reported have been normalized to account for the Au elec-trode area (to ensure comparability of the results). Normalized capacitance for DiffCap Buffer is shownin Figure 4.5. In addition, the influence of each component (e.g., phosphate, hydroxide, and chlorideions) on capacitance measurements is shown.4.3.4 Estimation of Surface CoverageFigure 4.6 shows a selection of capacitance–potential curves of increasing TMS-EDTA concentrationsin the DiffCap buffer. The curve labeled “Buffer” represents the capacitance of a Au bead electrode in694.3. Results and DiscussionFigure 4.2: (A) Raw capacitance data for DiffCap Buffer measurements from four independent experi-ments (from four different days). (B) Averaged raw DiffCap Buffer data and its corresponding standarddeviations.704.3. Results and DiscussionFigure 4.3: The stability of freshly cleaned bare Au bead electrode (circle) and TMS-EDTA-modified Aubead electrode (square) in DiffCap buffer at open-circuit potential (OCP).Figure 4.4: Raw capacitance data for three independent measurements (from three different days) of20 μM TMS-EDTA in DiffCap buffer.714.3. Results and DiscussionFigure 4.5: Averaged and normalized electrochemical differential capacitance data at equilibrium (360s with stirring) for individual DiffCap buffer components: (1) 100 mM phosphate (circle), (2) 100 mMphosphate and 65 mM KOH (square), and (3) DiffCap Buffer (triangle) – 100 mM phosphate, 65 mMKOH, and 500 mM KCl.DiffCap buffer in the absence of TMS-EDTA. In general, as the concentration of TMS-EDTA increased,the measured capacitance value decreased, due to TMS-EDTA adsorbing onto the Au electrode dis-placing water, chloride ions, and hydroxide ions. At a potential of −1.1 V, the capacitance values weresimilar (i.e., no significant decrease in capacitance) for the DiffCap buffer solution and for the solutionswith TMS-EDTA at concentrations of less than 50 mM. This result indicates high reproducibility of themeasurements and desorption of TMS-EDTA from the Au electrode at this potential. At TMS-EDTA con-centrations above 50 mM, pseudocapacitance features at −1.1 V (related to the kinetics of desorptionand adsorption process) were observed (see Section 2.4.2.4).208A constant minimum in capacitance was observed for potentials between −1 and 0.2 V at 200 mMTMS-EDTA concentration (see Figure 4.6). The minimum value of capacitance at each potential wasdetermined in order to calculate coverage (using eq 4.1). Assuming the curve for 200 mM TMS-EDTAcorresponds to a maximum surface coverage (θ=1), the fractional surface coverage (θ) as a functionof both potential (E) and TMS-EDTA concentration was calculated. For potentials less than −0.6 V,pseudocapacitance artifacts were observed (see Section 2.4.2.4); hence this range of potentials didnot fit the parallel plate model as stated in eq 4.1 (see Figure 4.7A) and these potentials were notanalyzed. Therefore, the surface coverage was calculated in a limited potential window between −0.5724.3. Results and DiscussionFigure 4.6: Capacitance–potential curves of increasing TMS-EDTA concentrations in DiffCap Buffer(select curves shown). For each concentration, capacitance was measured after each potential washeld for 6 min with stirring. Series resistor–capacitor equivalent circuit was assumed for calculations.and 0.2 V (Figure 4.7B). In general, as the concentration of TMS-EDTA increased, the calculated surfacecoverage increased. For potentials between −0.2 and 0 V, pseudocapacitance features due to Cl– orOH– adsorption (see Section 2.4.2.4) influenced the capacitance and the calculation of coverage (seeFigure 4.5).4.3.5 Determining the Gibbs Free Energies of TMS-EDTA Adsorption onto AuWith surface coverage data, the free energy of TMS-EDTA adsorption at each applied electrode potentialwas calculated by fitting to a Langmuir (dashed curve) or a Frumkin (solid curve) adsorption isotherm(see Figures B.11 to Figure B.16 of Appendix A). The results at selected potentials of −0.5 V (Figure4.8A), −0.25 V (Figure 4.8B), and 0.1 V (Figure 4.8C) are shown. It was observed that at negativeelectrode potentials both Langmuir and Frumkin isotherms described the data adequately. Surfaceexcess was likely smaller at more negative potentials since TMS-EDTA is negatively charged at highpH (see Figure A.1 of Appendix A). However, as the electrode potential became more positive, theLangmuir isotherm failed to adequately fit the experimental data. The surface excess may be greaterat more positive potentials resulting in increased repulsion among the adsorbed TMS-EDTA molecules,for which the Frumkin isotherm better describes the experimental data.734.3. Results and DiscussionFigure 4.7: (A) The fractional surface coverage curves calculated from capacitance values using Cθ =Cθ=0(1− θ)+Cθ=1θ. (B) A limited potential window is shown.744.3. Results and DiscussionFigure 4.8: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of (A) −0.5 V, (B) −0.25 V, and (C)0.1 V.754.3. Results and DiscussionPotential-dependent free energies of adsorption (Figure 4.9A) and lateral interaction parameters (Fig-ure 4.9B) derived from the Frumkin isotherm analysis are shown. Figure 4.9A reveals that the spontane-ity of TMS-EDTA adsorption on the Au electrode becomes more favorable as the applied potential (vsSCE) becomes more positive, similar to the adsorption of negatively charged chloride or hydroxide ionson Au.170,171 Potential-dependent free energies of adsorption were determined to be ∼ −20 to −30kJ/mol for the potential window shown (−0.5 to 0.2 V). A positive Frumkin lateral interaction parameterfrom Figure 4.9B indicates repulsion among adsorbed TMS-EDTA and that this repulsion increases asthe applied potential becomes more positive. These results can be explained as arising due to the sur-face excess likely being greater at more positive potentials, resulting in increased repulsion among thenegatively charged TMS-EDTA molecules. However, for potentials between −0.2 and 0 V, pseudoca-pacitance characteristics of the electrolytes were observed, and as a result, interpolated curves in thispotential window are also shown (measured data shown with dotted lines and interpolation shown withsolid lines in 4.9).Nevertheless, it should be noted that TMS-EDTA adsorptionwasmeasured using themulti-componentDiffCap buffer (100 mM phosphate, 500 mM KCl, and 65 mM KOH). The Gibbs free energy of adsorp-tion of hydroxide on the Au(111) electrode was found to be ∼ −110 to ∼ −120 kJ/mol from ∼ −0.1 to∼0.2 V, respectively.171 For chloride, the Gibbs free energy of adsorption at the Au(111) electrode wasfound to be ∼ −100 to ∼−110 kJ/mol within a similar potential range.170 Moreover, for hydroxide ions,adsorption occurs within the potential range from −0.4 to 0.2 V, limited by desorption (E<−0.4 V) andgold oxide formation (E>0.2 V).171 We also restricted our analysis to a similar potential range (between−0.5 to 0.2 V). However, within this potential range, the adsorption of TMS-EDTA is in competition withwater molecules, hydroxide ions, and chloride ions for the Au surface. In this complex system, adsorp-tion is a displacement process; therefore, the adsorption free energy of TMS-EDTA on Au is likely morefavorable than what the data presented here suggest. It is important to point out that the change inGibbs free energy depends on the adsorbate/adsorbent bond strength (enthalpy change) and entropychange (see Section 2.3). If the entropy change of this system is assumed to be negligible due to thehigh electrolyte concentrations in DiffCap buffer, then the apparent Gibbs free energies of adsorptionobtained may provide a rough estimate of the enthalpy change (i.e., strength of TMS-EDTA adsorptionon Au).Finally, recall that the typical wet chemical modification was achieved by immersing the Au substratein a 10% TMS-EDTA (v/v) solution with dissolved oxygen. The open-circuit potential (OCP) of the Ausubstrate under these conditions (see Figure 4.1) was ∼0.05 V (vs SCE). The free energy of adsorption764.3. Results and DiscussionFigure 4.9: Potential-dependent (A) free energies of adsorption and (B) lateral interaction parameter α,which are determined from Frumkin isotherm fitting. Positive α indicates repulsive interaction. Continu-ous interpolated values (solid line) are shown. Errors are estimated by robust fitting routine in MATLABand represent 95% confidence interval.774.4. Conclusionsat this deposition potential (obtained from Figure 4.9A) suggests that TMS-EDTA strongly modifies theAu surface. These results support the results that TMS-EDTA-modified Au could be used to capturestreptavidin and could withstand stringent washes (see Section 3.3). Also, at highly negative potentials(i.e., ∼ −1.1 V), TMS-EDTA adsorbs minimally onto Au (i.e., TMS-EDTA layer on Au may be electro-chemically removed).4.4 ConclusionsThermodynamic studies of TMS-EDTA adsorption on Au have been performed using electrochemicalmethods. Electrochemical differential capacitance measurements reveal that TMS-EDTA adsorption onAu is potential-dependent.For potentials between−0.5 to 0.2 V, the apparent Gibbs free energies of adsorptionwere determinedto be ∼ −20 to −30 kJ/mol in the complex electrolyte solution (measured under relevant conditions foruse in future sensor fabrication and biosensor applications). These results suggest that at more positivepotentials (i.e., at the potential typical for surface modification), adsorption is more thermodynamicallyfavorable. Due to its negative charge, there is more repulsive lateral interaction among the adsorbedTMS-EDTA at these potentials.At very negative potentials (∼ −1.1 V), there is minimal adsorption. Since the creation of a cleansurface is essential for electrochemical sensing applications, this result suggests that the TMS-EDTAlayer may be removed by applying negative potentials. The knowledge gained from these studies wasvaluable toward applying TMS-EDTA to construct robust PDMS-based electrochemical systems. Theseresults will be reported in the next chapter.78Chapter 5Fabrication of PDMS-BasedElectrochemical Cells UsingTMS-EDTA5.1 Synopsis2,4The fundamental studies of TMS-EDTA adsorption on Au (presented in Chapters 3 and 4) providedvaluable data for the rational design of integrated sensors and biosensors. The results suggested thatTMS-EDTA can potentially be used in a glass or gold-coated glass substrate for developing leak-freemicrofluidic devices. First, shear tests were conducted to determine the feasibility of using TMS-EDTAto chemically modify either glass or gold slides (i.e., to form terminal carboxyl groups) in order to enablebonding to polydimethylsiloxane (PDMS) slabs chemically modified with 3-APTMS (i.e., to form termi-nal primary amino groups). The strength of this carboxyl-amine bonding strategy was compared withother chemical bonding methods. Subsequently, the proposed carboxyl-amine bonding strategy wasimproved and refined for electrochemical applications. Pressure leak tests were conducted on deviceswith PDMS cell chambers bonded to 3-electrode substrates to obtain a more realistic measure of thebond strength under aqueous conditions. Finally, a method to electrochemically remove TMS-EDTAfrom the Au surface (inside a bonded PDMS-based device) was developed.795.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding Strategy5.2 Demonstrating the Feasibility of a Novel Carboxyl-AmineBonding Strategy5.2.1 Experimental Section5.2.1.1 MaterialsAll reagents were from commercial sources and used as-received without further purification: N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acid, Na salt (TMS-EDTA; 50% in water) was fromUnitedChemical Technologies (Bristol, PA); PDMS precursor and curing agent (Sylgard 184) were from DowCorning (Midland, MI); sodium hydroxide (NaOH; certified ACS) was from Fisher Scientific (Ottawa, ON,Canada); (3-aminopropyl)-trimethoxysilane (3-APTMS; 97%), (3-mercaptopropyl)-trimethoxysilane (3-MPTMS; 95%), 1,2-bis(trimethoxysilyl)ethane (BTMSE; 96%), 11-mercaptoundecanoic acid (11-MUA;95%), N-hydroxysuccinimide (NHS; 98%),N-(3-(Dimethylamino)propyl)-N´-ethylcarbodiimide hydrochlo-ride (EDC; 98%), acetone (ACS reagent, ≥99.5%), and 2-propanol (ACS reagent, ≥99.5%) were fromSigma-Aldrich (Oakville, ON, Canada); glass microscope slides were from VWR International (Missis-sauga, ON, Canada); nitrogen and argon (Ultra High Purity 5.0) were from Praxair Canada Inc. (Mis-sissauga, ON, Canada); and ethyl alcohol (denatured) was from UBC Zoology Stores (Vancouver, BC,Canada). For all studies, ultrapure water (∼18.2 MΩ-cm) produced by a Milli-Q water purification sys-tem (EMD Millipore) was used. The chemical structures of the molecules used in the shear tests arepresented in Figure 5.1.5.2.1.2 Fabrication of PDMS Slabs (Planar PDMS Surfaces)PDMS slabs with planar surfaces for shear tests were fabricated using standard soft lithography tech-niques (see Section 2.1.2) as shown in Figure 5.2. In short, a 10:1 (w/w) mixture of PDMS precursorand curing agent was prepared and poured onto a 5-inch aluminum weighing pan (VWR International;Mississauga, ON, Canada). After degassing, the pan was placed in an oven and PDMS was cured at60 °C for 2 hours. Prior to being used, the 6 mm thick PDMS was peeled off of the aluminum pan andcut into 25.4 mm by 12.7 mm pieces with razor blades.5.2.1.3 Fabrication of Gold-Coated Glass SubstratesThe glass slides were cleaned in a piranha solution (5:1 H2SO4:H2O2) at 80 °C for 15 min, rinsed withultrapure water, and dried in a nitrogen stream. The cleaned slides were subsequently placed into a 200805.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding StrategyFigure 5.1: Chemical structures of the molecules used in shear tests: 3-APTMS, 3-MPTMS, BTMSE,TMS-EDTA, and 11-MUA.Figure 5.2: Scheme for producing PDMS slabs using soft lithography. (A) A flat aluminum weighing panis used as the mold. (B) Liquid PDMS is poured onto the mold. (C) Mold with liquid PDMS is baked inan oven. (D) Solidified PDMS is peeled off and (E) cut into slabs of appropriate sizes.815.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding Strategy°C oven for 15 min to eliminate residual moisture. A 20 nm chromium adhesion layer and a 200 nm goldlayer were sequentially deposited onto clean glass slides using an electron-beam evaporator (AMPELAdvanced Nanofabrication Facility at UBC Vancouver).5.2.1.4 Cleaning of Glass or Gold-Coated Glass SubstratesThe glass slides were cleaned by sequential immersions in acetone, 2-propanol and ultrapure water (with10-min sonication for each). The gold-coated glass slides were cleaned by sequential immersions inacetone, 2-propanol, and ultrapure water (repeated three times, without sonication). Cleaned substrateswere blown dry in a nitrogen stream.5.2.1.5 Surface Functionalization of Glass or Gold-Coated Glass SubstrateThe surface of the PDMS was modified with an amino-silane to create a layer of reactive primary aminogroups. Briefly, PDMS slabs were exposed to UV-ozone (UVO-Cleaner, model 42, Jelight Co. Inc., CA)for 5 min and immediately immersed into a 10% (v/v) solution of 3-APTMS in ethanol for 1 h. The PDMSslabs were then washed three times in ethanol, followed by three times in ultrapure water, and then driedunder N2for immediate use. Glass and gold substrates were cleaned as previously described, followedby exposing to UV-ozone for 5 min. Clean glass substrates were immersed in aqueous solution of 10%TMS-EDTA (v/v) or ethanolic solution of 10% BTMSE (v/v), whereas gold substrates were immersed ineither ethanolic solutions of 10% 3-MPTMS (v/v) or 10 mM 11-MUA, or aqueous solution of 10% TMS-EDTA (v/v) for 2 h at room temperature. All substrates, except for TMS-EDTA, were washed three timesin ethanol and three times in ultrapure water, and then dried under a N2stream. TMS-EDTA-modifiedsubstrates were washed three times in ultrapure water and dried under a N2stream.5.2.1.6 Irreversible BondingCarboxyl-terminated substrates functionalized with TMS-EDTA or 11-MUA were subjected to carbodi-imide activation (50 mM NHS and 200 mM EDC for 30 min), followed by drying under N2. Thesesubstrates were then placed in contact with PDMS surfaces functionalized with 3-APTMS. Silanol-terminated surfaces (i.e., BTMSE, 3-MPTMS) were hydrolyzed for 1 h in a 1 M NaOH solution (seeSection 2.1.1.2), followed by rinsing in ultrapure water and drying under N2. These hydrolyzed surfaceswere placed in contact with unmodified PDMS surfaces that were exposed to UV-ozone for 5 min. Con-trol experiments consisted of bringing together the surfaces of both unmodified PDMS and clean glass825.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding Strategyslides that were exposed to either UV-ozone or oxygen-plasma for 5 min. For all bonding methods, theglass or gold slides were immediately brought into contact with the PDMS surfaces and allowed to bondat room temperature for 1 h.5.2.1.7 Shear Tests of PDMS Slabs Bonded to Glass or Gold-Coated Glass SubstratesIn total, four types of PDMS−glass bonding strategies and three types of PDMS−gold bonding strategieswere analyzed. The strengths of various bonding strategies were measured by performing shear tests42(Material Testing System, MTS 810, MTS Systems Corporation; Eden Prairie, MN) with the followingprocedural modifications (see Figure C.1 of Appendix C). The back of each PDMS-bonded substrate(i.e., the glass slide that was not bonded to PDMS) was epoxy-glued to a wood block, which was thenclamped to a testing stage. In each test, the load cell was aligned and the PDMS slab was pushed off ofthe solid substrate (at a rate of 1 mm/min). The continuous displacement of the load cell applied a shearforce on the PDMS material. The force curve was recorded against elapsed time (four measurementsper second). The peak of each force curve represents the point of failure for the PDMS−substratebond. The bond strength of each sample was calculated as the average of the highest force values ina 2-second time interval. Three elastic failure curves (from three independent samples) were obtainedfor each of the bonding method, and standard deviation was calculated using these values. The forcevalues were then converted to pressures by dividing the PDMS−substrate contact area (see Tables C.1and C.2 of Appendix C).5.2.2 Results and DiscussionPDMS slabs were modified with ethanolic 3-APTMS in order to form terminal primary amino groups.Water contact angle measurements using the sessile drop method were performed on the surfaces ofbare PDMS (Figure 5.3A), PDMS exposed to UV-ozone (UVO) for 5 min (Figure 5.3B), and 3-APTMS-modified PDMS (Figure 5.3C) to confirm the surface modification. A single measurement was performedon each sample for proof-of-concept. The PDMS surface modified with 3-APTMS showed a decrease inthe water contact angle, which indicated that the surface became more hydrophilic than the bare PDMSand the UVO PDMS, confirming chemical functionalization (see Section 2.2.1).To illustrate the concept of the carboxyl-amine strategy, a detailed scheme of the bonding method for11-MUA-modified gold is presented in Figure 5.4. In short, PDMS surface was modified with 3-APTMSto form terminal primary amino groups (Figure 5.4A). Gold-coated glass (or simply gold) substrate was835.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding StrategyFigure 5.3: Water contact angle measurements on (A) bare PDMS, (B) PDMS exposed to UV-ozone(UVO) for 5 min, and (C) 3-APTMS-modified PDMS.845.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding Strategymodified with 11-MUA to form terminal carboxylic acid groups (Figure 5.4B). Finally, the terminal car-boxylic acid groups on Au were treated with carbodiimide activation (i.e., NHS/EDC), followed by react-ing with 3-APTMS-modified PDMS and allowing to form a bond by physical contact at room temperature(Figure 5.4C). The glass or gold substrates modified with TMS-EDTA were also treated with the sameprocedures.For shear tests, gold substratesweremodified with 11-MUA or TMS-EDTA, and glass substratesweremodified with TMS-EDTA in order to create terminal carboxyl groups. Following carbodiimide activation,the solid (gold or glass) substrates were brought into contact with PDMS slabs modified with 3-APTMSafter ∼1 s. Strong bonding was observed after 1 h. A few other chemical bonding strategies werealso studied for comparison. Glass substrates modified with BTMSE or gold substrates modified with3-MPTMS were treated with 1 M NaOH for 1 h prior to bonding with UV-ozone-treated PDMS via silanecross-linking. Clean glass and unmodified PDMS treated with UV-ozone and oxygen-plasma were alsobonded together, serving as controls.The bond failure curves for TMS-EDTA-modified glass and 3-APTMS-modified PDMS are presentedin Figure 5.5A. The peak value of each shear test curve represents the failure point of the PDMS−substratebond. The small variation observed in the three replicates measured illustrates the reproducibility ofthis characterization method. The bond strengths of four different glass−PDMS bonding strategies arepresented in Figure 5.5B. Figure 5.5C shows the results from three different gold−PDMS bonding strate-gies. In all instances, the bond failed at the PDMS−substrate interface by delamination due to the appliedshear stress. However, it was difficult to determine the different modes of failure from these shear test re-sults. Nevertheless, a thin layer of PDMS residue was left on the substrate, confirming the strong bondsformed. The pressure was calculated by dividing the peak force value (determined from the shear test)by the PDMS−substrate contact area (see Tables C.1 and C.2 of Appendix C).To determine the validity of the shear test results, UV-ozone and oxygen-plasma bond strength val-ues were compared with values reported in the literature for PDMS−PDMS devices. Typically, bondingmethods using oxygen-plasma result in stronger bond strengths than those using UV-ozone.209 The re-sults presented in Figure 5.5B follow this trend for PDMS−glass devices. In addition, the oxygen-plasmabond strength value is in close agreement with previously published values for PDMS−PDMS bondeddevices.210 This result is expected due to the siloxane bonds formed (see Section 2.1.2).From Figure 5.5, it was observed that the carboxyl-amine bonding strategy produced bond strengthsthat were comparable to those of conventional methods using UV-ozone treatment for glass substratesand 3-MPTMS treatments for gold substrates. The BTMSE-bonded PDMS−glass structure produced a855.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding StrategyFigure 5.4: (A) Modification of PDMS with 3-APTMS to form primary amines. (B) Modification of goldwith 11-MUA to form carboxylic acids. (C) Carbodiimide activation of the carboxylic acid groups, followedby irreversible bonding of PDMS to gold by physical contact at room temperature.865.2. Demonstrating the Feasibility of a Novel Carboxyl-Amine Bonding StrategyFigure 5.5: (A) Bond failure curves of TMS-EDTA-modified glass and 3-APTMS-modified PDMS slabs.The peak of each curve represents the failure point of the PDMS−substrate bond. (B) Bond strengths offour different PDMS−glass bonding strategies: UV-ozone treatment of both glass and PDMS surfaces(UVO); oxygen-plasma treatment of both glass and PDMS surfaces (O2Plasma); TMS-EDTA-modifiedglass with 3-APTMS-modified PDMS (TMS-EDTA & 3-APTMS); and BTMSE-modified glass with UV-ozone-treated PDMS (BTMSE & UVO). (C) Bond strengths of three different PDMS−gold bonding strate-gies: 3-MPTMS-modified gold with UV-ozone-treated PDMS (3-MPTMS& UVO); 11-MUA-modified goldwith 3-APTMS-modified PDMS (11-MUA & 3-APTMS); and TMS-EDTA-modified gold with 3-APTMS-modified PDMS (TMS-EDTA & 3-APTMS). All samples were analyzed in triplicate. Error bars representstandard deviation.875.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cellscomparable bond strength as oxygen-plasma-bonded glass and PDMS. The bonding method that uti-lized BTMSE did not employ the carboxyl-amine bonding strategy outlined above, but instead served asanother demonstration of the solution-based chemical modification of glass for PDMS bonding. Addi-tional control experiments were conducted by manual peel tests to determine the importance of formingboth terminal carboxyl groups and terminal primary amino groups on solid substrate (i.e., glass or gold)and PDMS, respectively (see Table C.3 of Appendix C). The bonded structures were then completelyimmersed in water to determine its hydrolytic stability, and started to show signs of weakening after 2weeks. It is known that for a short immersion time (i.e., 24 h), PDMS does not swell significantly in wa-ter.52 However, it has been reported that for a longer immersion time (i.e., 2 weeks), PDMS does indeedswell in water.211 The PDMS swelling may have caused the detachment of PDMS slabs from solid goldor glass substrates. As a result, it is difficult to determine the hydrolytic stability of the bonds betweenthe PDMS slabs and the substrates over time. Nevertheless, devices kept at ambient conditions in airhas not shown signs of weakening since 2009.The carboxyl-amine bonding method presented here is similar to the approach developed by Leeand Chung, which employed epoxy-amine chemistry to bond PDMS−PDMS devices.212 However, thesubstrates reported here were modified with silanes or alkanethiols terminated with carboxyl groups, asopposed to epoxy groups. As a result, more specific covalent coupling with free primary amino groupscan be achieved,213 because epoxy is known to react with other functional groups including alcohols,carboxylic acids, and acid anhydrides.214 Nevertheless, the important result obtained from the sheartests presented here was that TMS-EDTA could chemically modify either glass or gold substrates forcarboxyl-amine bonding with PDMS modified with terminal primary amino groups. These results (andthe results from SPR experiments in Section 3.3) suggest that TMS-EDTA-modified Au has at least somecarboxyl groups available for reaction that are oriented away from the Au surface, resulting in irreversiblecoupling with substances containing terminal primary amino groups.5.3 Applying TMS-EDTA to Construct Leak-Free PDMS-BasedElectrochemical CellsAs mentioned in Section 2.5, there is a growing interest in creating microfluidic electrochemical sensorsdue to their inexpensive experimental setup and suitability for miniaturization.108 Typically, Au electrodesare sputtered onto a flat glass slide, and a PDMS chip is placed in contact with the solid substrate, form-885.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cellsing enclosed channels and/or cell chambers (see Figure 1.2 of Section 1.1).186 There are two separateand well-defined interfaces: PDMS−glass and PDMS−Au interfaces. It is simple to form either a strongPDMS−glass bond (e.g., using UV-ozone or oxygen-plasma) or a strong PDMS-Au bond (e.g., using 3-MPTMS). However, it is difficult to form both PDMS−glass and PDMS−gold bonds at the same time usingconventional chemical bonding strategies. Therefore, aqueous solution may leak at the PDMS−Au inter-face with just a slight increase in fluidic pressure (e.g., during loading of aqueous electrolyte) when thegold surface is not chemically modified.57 It is important for these PDMS-based electrochemical devicesto form a tight seal because the leakage problem will negatively impact electrochemical measurementsand data analysis (see Section 2.5.1).In Section 5.2, the shear tests were conducted on samples of either glass or gold substrates bondedto PDMS slabs. The positive results indicated the feasibility of bonding a substratewith both glass and Ausurfaces (i.e., a mixed Au/glass substrate) to PDMS using the same carboxyl-amine chemistry with TMS-EDTA. The rest of this chapter is focused on the application of TMS-EDTA to develop leak-free PDMS-based electrochemical cell with a 3-electrode substrate (see Figure 5.6). Since most electrochemicalsensing applications require the use and transfer of aqueous solutions, it is important to determine themaximum fluidic pressure that these bonded devices can withstand.215 Pressure leak tests with dyedwater were conducted on devices bonded using the carboxyl-amine strategy (with the oxygen-plasmabonding method serving as a control). Finally, various electrochemical surface cleaning methods weretested to determine the best protocol that could remotely desorb TMS-EDTA from the Au surface withina bonded device.Almost all aspects of the device fabrication process have been refined and improved in order toconstruct robust PDMS-based electrochemical cells. Detailed experimental procedures are summarizedbelow, and some of the most important procedural modifications are highlighted here: (1) the PDMSprecursor and curing agent was thoroughly mixed by hand and machine (instead of only by hand), (2)Ti (instead of Cr) was used as the adhesion layer between glass and gold, (3) a modified RCA cleaningmethod (instead of acetone, 2-propanol and ultrapure water) was used to clean all solid substrates (i.e.,glass, gold-coated glass, and 3-electrode substrates), and (4) aqueous (instead of ethanolic) 3-APTMSwas used to reduce PDMS swelling.895.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.6: Schematic of the fabricated 3-electrode device. (a) Top view of the PDMS-bonded device. (b)Cross-sectional view of the device through the center (i.e., down the length) of the device. (c) Enlargedview of the proposed carboxyl-amine chemistry between PDMS and the gold/glass substrate.905.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cells5.3.1 Experimental Section5.3.1.1 Fabrication of PDMS Cell Chambers3D-Printed Molds Initially, 3D-printed molds were used to create the PDMS cell chambers with cylin-drical side-walls due to its low cost and fast prototyping. The 3D-printed molds were designed in Solid-works. The designs were exported as .stl files and printed using the 3D printers in the Electrical andComputer Engineering Lightning Lab (UBC Vancouver). All molds were printed with the cell chambersin a positive-relief format (i.e. a negative of the actual chamber) and made of an acrylic-based plastic(VeroWhitePlus, Objet, Inc., MA). However, the PDMS structures created from the 3D-printed molds re-sulted in rough PDMS surfaces (since the surfaces of 3D-printed molds were uneven). As a result, thePDMS cell chambers with smooth surfaces were fabricated using a 2-step process as described below.Step 1: Creating PDMS Cell Chambers with the 3D-Printed Molds Using the 3D-printed molds,the initial PDMS cell chambers were fabricated with the procedure described in Figure 5.7. In short, a10:1 (w/w) mixture of PDMS precursor and curing agent was first mixed by hand and then by machine(Thinky AR-250, THINKY USA Inc., CA). The 3D-printed molds were surrounded with single-sided tape(3M Canada) on the rectangular perimeter in order to create sidewalls and to contain uncured liquidPDMS. Uncured PDMS mixture was then poured onto the 3D-printed molds and placed in a vacuumchamber to remove air bubbles. After degassing, the chambers were cured in an oven for 3.5 hours at60-65 °C. Afterwards, the tape was removed and the cured PDMS structures were peeled off of the 3D-printed molds. The surface of the 3D-printed mold was rough, and as a result, the surface of the PDMScell chamber produced was also rough. After making the initial PDMS cell chambers, the 3D-printedmolds could be re-used by rinsing in 2-propanol, ethanol, and ultrapure water (repeated three times),and drying with in-house nitrogen gas.Step 2: Creating PDMS Cell Chambers with Glass/PDMS Molds Figure 5.8 details the steps usedto create the final PDMS cell chambers with new glass/PDMS molds. First, additional mixed and un-cured PDMS was poured into the hollow chamber of a PDMS cell chamber (made using the protocoldescribed in Section 5.3.1.1), which served as the negative mold in order to fabricate a PDMS disk witha volume of ∼50 μL. After degassing and baking the PDMS as before, the disks were removed from thePDMS negative molds. The solid PDMS disks were bonded to clean glass slides by air-plasma (PlasmaCleaner, model PDC-001, Harrick Plasma; Ithaca, NY) in order to create the final positive molds. Again,tape was used to create sidewalls around the PDMS/glass positive molds. Newly mixed and uncured915.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.7: Step 1 of 2: Scheme for producing PDMS cell chambers from 3D-printed molds. (A) A3D-printed mold is made with desired features. (B) Liquid PDMS is poured onto the mold. (C) Moldwith liquid PDMS is cured. (D) Solidified PDMS is peeled off. The rough surface of the 3D-printed moldresulted in a rough surface on the PDMS cell chamber produced.PDMS was poured into these molds, degassed in a vacuum chamber, and baked as before. SolidifiedPDMS cell chambers were peeled off of the molds, and cut with razor blades into appropriate sizes.Similarly, the PDMS/glass molds could be re-used by rinsing in acetone and cleaning using the samemethod as the 3D-printed molds (described above).Mold Designs A few of the 3D-printed mold designs are shown in Figure 5.9A, and the glass/PDMSmold design is shown in Figure 5.9B. PDMS cell chambers with smooth surfaces were fabricated withthese two types of molds using the 2-step process as described above. Each individual cell chamberwas designed to have a radius of 3.8 mm and a height of 1.2 mm (i.e., a volume of ∼50 μL). Each PDMScell chamber was hole-punched with a 0.75 mm PDMS biopsy hole puncher for solution injection.5.3.1.2 Fabrication of Gold-Coated Glass SubstratesGlass slides were first rinsed with acetone, 2-propanol and ultrapure water, and then cleaned in a piranhasolution (5:1 H2SO4:H2O2) at 80 °C for 15 min, followed by rinsing with ultrapure water and drying in anitrogen stream. The cleaned slides were subsequently placed into a 120 °C oven for 5 min to eliminateresidual moisture. A 10 nm Ti adhesion layer and a 200 nm gold layer were sequentially deposited ontoclean glass slides using the electron-beam evaporator as described in Section 5.2.1.3.5.3.1.3 Fabrication of the 3-Electrode SubstratesThe 3-electrode substrate used Borofloat 33 as the glass wafer (4-inch diameter and 500 μm thickness,University Wafer; Boston, MA), Ti as the metal adhesion layer, and Au as the metal for the electrode.925.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.8: Step 2 of 2: Scheme for producing the final PDMS cell chambers using glass/PDMS molds.(A) PDMS cell chamber made with the 3D-printed mold is used as the initial mold. (B) Liquid PDMSis poured into the hollow chamber. (C) Mold with liquid PDMS is cured and a small disk is produced.(D) Small disk is bonded to a clean glass slide by air-plasma. (E) Liquid PDMS is poured onto the newglass/PDMS mold. (F) Mold with liquid PDMS is baked. (G) Solidified PDMS cell chamber is peeled offand cut into appropriate sizes.935.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.9: Molds used for creating PDMS cell chambers with smooth surfaces. (A) Designs of the3D-printed molds for creating the initial PDMS cell chambers (Step 1), from which the small PDMS diskswere created. (B) Small PDMS disks were then bonded to clean glass slides by air-plasma in order tocreate the final mold design. The final PDMS cell chambers were fabricated using these PDMS/glassmolds (Step 2).945.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.10: Design of the 3-electrode substrate.These substrates were fabricated by photolithographically patterning a photoresist mask, depositing10 nm Ti adhesion layer and 200 nm Au onto the glass wafer, and lifting off the sacrificial photoresistlayer.4 The glass wafers were then diced into smaller (25.78 mm length and 10.8 mm width) 3-electrodesubstrates (see Figure 5.10). The Au electrode lead(s) had a width of either 770 μm or 500 μm.5.3.1.4 Modified RCA Cleaning MethodSince gold and glass surfaces are present on the 3-electrode substrate, an adequate cleaning methodmust be used to simultaneously clean both types of surfaces. The protocol described here is based ona previously published cleaning/etching method.216 The glass, gold-coated glass or 3-electrode sub-strates were first cleaned by sequential immersions in acetone, 2-propanol, methanol and ultrapurewater (with 15-min sonication for each). The substrate was then immersed for 15 min in a solution mix-ture called RCA (50 mL ultrapure water, 10 mL ammonium hydroxide, and 10 mL hydrogen peroxide).Methanol (certified ACS), ammonium hydroxide (reagent ACS), and hydrogen peroxide (certified ACS)were from Fisher Scientific (Ottawa, ON, Canada). After briefly rinsing the substrates with ultrapurewater, they were placed in a 0.1 M nitric acid solution for 30 min. Nitric acid (certified ACS; 69-70%)was from VWR BDH Chemicals (Mississauga, ON, Canada). The cleaned substrates were rinsed withultrapure water and blown dry in an argon stream.5.3.1.5 Carboxyl-Amine Bonding Strategy Using TMS-EDTAThe carboxyl-amine bonding strategy using TMS-EDTA has been developed in Section 5.2. To avoidany ambiguity, the refined experimental procedures are summarized and reproduced here. All of thePDMS-based devices were fabricated using this protocol.955.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFormingTerminal Carboxyl GroupsonGlass, Gold-CoatedGlass or 3-Electrode Substrate Cleanedglass, gold-coated glass or 3-electrode substrate was exposed to UV-ozone for 5 min. The substrateswere then immersed in an aqueous solution of 10% TMS-EDTA (v/v) for 2 h at room temperature. TMS-EDTA-modified substrates were washed three times in ultrapure water and dried under an argon stream.Carbodiimide Activation Aqueous solutions of 100 mM N-hydroxysuccinimide (NHS) and 400 mMN-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC) were prepared separately. Anaqueous mixture of these two solutions (volume ratio of 1:1) was immediately prepared prior to use,in order to make a solution with a final concentration of 50 mM NHS and 200 mM EDC. This mixturewas used to activate the terminal carboxyl groups on solid substrates.Forming Terminal Primary Amino Groups on PDMS The surface of the PDMS was chemicallymodified with (3-aminopropyl)-trimethoxysilane (3-APTMS) in order to create a layer of terminal primaryamino groups. First, the PDMS surface was cleaned using single-sided tape and exposed to UV-Ozonefor 5 min. Then, the PDMS surfaces were immediately immersed into a 10% (v/v) solution of 3-APTMSin ultrapure water for 1 h. The PDMS surface was then washed three times with water and dried withargon for immediate use.Carboxyl-Amine Bonding Solid substrates modified with TMS-EDTA were subjected to carbodiimideactivation for 30 min, followed by a quick rinse with ultrapure water and dried with argon. The activatedterminal carboxyl groups on the solid substrates were immediately brought into contact with terminalprimary amino groups on 3-APTMS-modified PDMS surfaces and allowed to react at room temperaturefor 1 h (see Section 2.1.3.1).5.3.1.6 Pressure Leak TestsTwo Types of PDMS Bonding Strategies The solid substrates (i.e., glass, gold-coated glass or 3-electrode substrates) were cleaned using the modified RCA cleaning method as described in Section5.3.1.4. Two types of PDMS bonding methods were tested by conducting pressure leak tests: (1)carboxyl-amine strategy and (2) air-plasma strategy. First, PDMS cell chambers were bonded to the solidsubstrates using the carboxyl-amine bonding strategy as described in Section 5.3.1.5. After bringing themodified PDMS and the activated solid substrates into contact, irreversible bonding was observed within1 h, but the pressure leak tests were conducted after 16 h. Second, the surfaces of both unmodified965.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsPDMS and clean solid substrate (i.e., glass, gold-coated glass or 3-electrode substrate) were exposedto air-plasma for 1 min and 15 sec and brought into contact to serve as control experiments.Supplies for Pressure Leak Tests All supplies were from commercial sources: green food-gradecolour dye was fromMcCormick Canada (London, ON, Canada); plastic syringes (3mL and 10mL) werefrom Becton Dickinson (Mississauga, ON, Canada); dispensing tips (straight and right-angled 21 gaugeand 22 gauge) with Luer lock attachment point were from Nordson EFD (East Providence, RI); plastictubing (Tygon R-3603) was from Sigma-Aldrich (Oakville, ON, Canada); plastic (Tygon)microbore tubing(0.51 mm ID x 1.52 mm OD) was from Cole-Parmer Canada (Montreal, QC, Canada); and pressuresensor (HDIB002GUSMD8P5) was from First Sensor/Sensortechnics (Mansfield, MA).Experimental Setup The experimental setup for pressure leak tests is shown in Figure 5.11. A 2megapixel digital colour camera (A3250U, OMAX) with a 0.5X reduction (A3RDF50, OMAX) lens wasmounted on a stereo zoom microscope (SZ6045, Olympus) and connected to a computer. OMAXToupView 3.7 was used to record images and videos of the pressure leak tests. Open-source hard-ware, Arduino ATmega328, was used to process the pressure sensor information and send the datato the computer. After placing the sample on the sample stage of the microscope, the syringe pump(KDS230, KD Scientific; Holliston, MA) was used to apply a continuous pressure to the syringe, whichin turn applied a pressure to the solution-filled PDMS cell chamber.The PDMS cell chamber of each sample was manually loaded by injecting green dye solution intothe inlet at a low pressure. A 3-way valve was used to connect the 10 mL syringe and the pressuresensor with Tygon tubing. A third tubing (connecting the 3-way valve to the cell chamber inlet) wasinitially purged of air with ultrapure water using the syringe. Subsequently, the outlet of the cell chamberwas blocked with a plug, while the inlet was connected to the 3-way valve by the third tubing (i.e., thewater-filled tubing). After placing the connected device under the microscope, the syringe pump pushedultrapure water from the syringe into the solution-filled cell chamber at a flow rate of 1 mL/min (whilethe pressure was monitored by the pressure sensor) until a leakage of the green dye was observed orthe maximum pressure measurable was achieved. For samples that leaked, the pressure was recordedat the first sign of leakage (e.g., when the green dye emerges at either the PDMS−glass or PDMS−Auinterface) as determined from the video images.975.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsSyringePumpComputerStereoMicroscopeArduinoATmega32810 mL Syringe3 Way ValveCameraPressure SensorFigure 5.11: Setup used to evaluate the strength of the bond (via leak pressure tests) between PDMSand solid substrates (glass, gold, and mixed Au/glass substrates).985.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cells5.3.1.7 Electrodesorption of TMS-EDTA from GoldSupplies for Electrodesorption A commercial Ag/AgCl reference electrode (RE-5B) from BASi Inc.(West Lafayette, IN) was used as the silver/silver chloride electrode (SSCE) RE. The in-house fabricatedAg/AgCl RE was calibrated against the commercial SSCE RE. Hexaamminecobalt(III) chloride (CoHex)and sulfuric acid (ACS, 95.0-98.0%) were from Sigma-Aldrich (Oakville, ON, Canada). Potassium chlo-ride and potassium hydroxide were from Fisher Scientific (Ottawa, ON, Canada).Instrumentation Measurements were taken using either a PGSTAT30 potentiostat, a PGSTAT12 po-tentiostat, or a μAutolab potentiostat. NOVA (versions 1.10 and 1.11) software (Metrohm Autolab;Utrecht, Netherlands) was used to control the potentiostats, collect the data, and perform the dataanalysis. First, the PDMS-based 3-electrode device was bonded using the carboxyl-amine strategyas described in Section 5.3.1.5. The inlet of the PDMS cell chamber on the device was connectedwith tubing to the working solution of interest. The outlet tubing was initially connected to a syringe.By applying a suction pressure with the syringe, the working solution was drawn into the cell chamber.Subsequently, the outlet tubing was plugged with the in-house fabricated Ag/AgCl RE and filled withsaturated chloride. The device was then connected to the potentiostat by DropSens µSTAT Cable Con-nector (DropSens; Asturias Spain). After inserting the connected device into a glass container, the glassopening was sealed by the Cable Connector. The container with the suspended PDMS-based devicewas deoxygenated with argon.Electrodesorption Themethod to electrochemically desorb TMS-EDTA from the Au surface inside thePDMS-bonded 3-electrode device was developed. The cell chamber was initially filled with a solution of750 μM CoHex (in a deoxygenated 100 mM phosphate buffer without chloride) and cyclic voltammetry(CV) measurements were conducted in an applied potential window of −0.5 V and +0.2 V. Then, thecell chamber was filled with 100 mM KOH (pH ∼13). The electrode potential was held at −1.4 V for 0.5sec, followed by holding the potential at open-circuit potential for 10 sec (repeated 200 times). The cellchamber was re-filled with the CoHex solution and CV was measured as before. Next, the cell chamberwas filled with 1 M sulfuric acid and the potential was cycled between−0.15 V and +1.7 V. Finally, a CVof the CoHex solution was measured again.For control experiments, a clean gold bead electrode was modified with aqueous 10% (v/v) TMS-EDTA solution for 2 h. Platinum was the CE and the commercial SSCE was the RE. The same surfacecleaning protocols (i.e., pulsing to a negative potential in KOH and cycling in sulfuric acid) were applied.995.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsSubsequently, the gold bead electrode was flamed with a butane torch and rinsed. CV measurementsof the CoHex solution were conducted before and after each of the three cleaning protocols as before.5.3.2 Results and Discussion5.3.2.1 Pressure Leak Tests with Bonded PDMS Cell ChambersPressure leak tests with PDMS cell chambers bonded to glass, gold, or mixed Au/glass substrates wereconducted (an example of the experimental result is shown in Figure C.2 of Appendix C). The pressuresensor can detect up to 200 kPa (i.e., 2 bars or ∼29 psi) of internal pressure (gauge) in the cell chamber.Prior to conducting the pressure leak tests, a hydrostatic pressure of 21 kPa was recorded. All of thepressures reported below (for samples that survived loading) have been subtracted by this value. Table5.1 summarizes the results from these tests for air-plasma and carboxyl-amine bonding strategies.Table 5.1: Results from pressure leak tests for air-plasma and carboxyl-amine bonding strategies, withthe internal pressure (gauge) of the cell chamber shown. Three replicates were performed for eachbonding strategy. Standard deviations are reported.PDMS Bonding Strategy Glass Substrate Au Substrate Mixed Au/Glass SubstrateAir-Plasma >179 kPa On Loading 8±6 kPaCarboxyl-Amine 146±6 kPa 38±9 kPa 50±5 kPaFor the air-plasma bonding strategy, the PDMS−glass devices did not leak (i.e., its leak pressuresurpassed the pressure sensor’s detection limit). On the other hand, the bare Au substrates cleaned withair-plasma did not bond with plasma-treated PDMS chambers; solution started to leak upon loading (∼3kPa). For PDMS bonded to the substrate with both Au/glass, the solution only leaked via the PDMS−goldinterface at the pressure of 8±6 kPa. These results are expected since the air-plasma method does notform a strong bond between PDMS and Au. Large variation in the measurements was observed due toweak PDMS−Au bond upon solution loading.For the carboxyl-amine bonding strategy, PDMS−glass devices showed a leak pressure of 146±6kPa. On the other hand, PDMS−gold devices showed a leak pressure of 38±9 kPa. For the mixedAu/glass substrate, solution also leaked via the PDMS−Au interface and a leak pressure of 50±5 kPawas obtained.In general, the air-plasma bonding strategy creates a very strong bond for the PDMS−glass interface,but the bonding for the PDMS−gold interface is extremely weak. On the other hand, the carboxyl-aminebonding strategy creates a relatively weaker bond for the PDMS−glass interface, but the bonding forthe PDMS−gold interface is improved. For the mixed Au/glass substrate, the carboxyl-amine bonding1005.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.12: (A) PDMS cell chamber bonded to the mixed Au/glass substrate using the air-plasmabonding strategy right after loading (i.e., at atmospheric pressure). (B) Initial leak pressure of the plasma-bonded device (i.e., 8 kPa). (C) Complete failure of the plasma-bonded device (i.e., 29 kPa). PDMSbonded to the 3-electrode substrate using the carboxyl-amine bonding strategy at (D) the atmosphericpressure after loading, (E) the leak pressure of plasma-bonded device (i.e., 8 kPa), and (F) the pressureof an initial solution leak (i.e., 49 kPa). Diameter of the cell chamber was 7.6 mm.strategy creates a device that can withstand a fluidic pressure∼6X stronger than the air-plasma bondingstrategy. Since the fluid pressure in microfluidic systems rarely exceeds ∼5 psi (∼34 kPa),217 the sealobtained with this carboxyl-amine bonding strategy is adequate for most microfluidic applications (butmay not be suitable for systems requiring high pressures).Some representative images from the pressure leak tests of the 3-electrode devices are shown inFigure 5.12. Figures 5.12A and 5.12D show that upon loading of the solution, no leaks were observedfor both plasma- and carboxyl-amine-bonded devices. At 8 kPa, the PDMS−gold interface of air-plasma-bonded device started to leak (Figure 5.12B), and the device failed completely at 29 kPa (Figure 5.12C).In contrast, the carboxyl-amine-bonded device remained leak-free (Figure 5.12E) at 8 kPa, and showedthe indication of a leak at 49 kPa (Figure 5.12F). Figure 5.12F also shows the bulging of the PDMS cellchamber due to the increased fluidic pressure. These results demonstrate that the 3-electrode devicesbonded using the carboxyl-amine strategy can create a strong seal around the PDMS cell chamber,primarily due to the increased PDMS−Au interaction.1015.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cells5.3.2.2 Electrodesorption of TMS-EDTA from GoldIn Section 5.3.2.1, it has been demonstrated that the carboxyl-amine bonding strategy (i.e., using TMS-EDTA on mixed Au/glass substrate) can create a seal around the PDMS cell chamber. It is also evidentthat TMS-EDTA remains on the Au electrode after bonding (see Figure 1.2C). This layer may be suit-able for the subsequent immobilization of biomolecules, as demonstrated in Section 3.3, for biosensingapplications. However, this layer may also compromise electrochemical measurements. For example,the TMS-EDTA layer may impede the electron transfer between electroactive species in solution andthe electrode surface. Consequently, a reproducible surface cleaning method needs to be developed inorder to remotely remove the adsorbed TMS-EDTA and to prepare a clean Au surface (in the device afterbonding) for effective electrochemical sensing. First, the electrochemistry of partially blocked electrodesneeds to be reviewed.Electrochemistry of Partially Blocked Electrodes By adding a redox active species to the elec-trolyte, the quality of an adsorbed layer on the electrode surface can be indirectly inferred. For example,a bare electrode will show the faradaic currents that correspond to the oxidation or reduction of the par-ticular species on the whole surface. When the electrode is covered with adsorbate molecules (e.g., anorganic layer), this layer may act as a barrier against the electron transfer (redox) of the electroactivespecies. Two results can occur: total blocking or partial blocking (i.e., slowing down) of the electrontransfer. This approach is commonly used as a general indicator of the cleanliness of the electrodesurface.218For example, the oxidation and reduction of hexacyanoferrate(III) were studied by linear potentialsweep and cyclic voltammetry (CV) on Au disk electrodes partially covered with photoresist.219 It wasassumed that some sites were active and that some sites were inactive. It was observed that at a higherdegree of coverage of the inactive photoresist (from θ =0.552 to 0.815), the separation between theoxidation and reduction peak potentials was increased. Subsequently, a model for charge transfer atpartially blocked surfaces was proposed.220 It was found that when the fractional coverage was small(i.e., not too close to maximum coverage), the electrochemical response was the same as the one foran unblocked electrode, but with a decreased apparent rate constant of electron transfer. The apparentelectron transfer slowed when the coverage increased. The current peak in the CV also decreasedwith a further increase in adsorbate coverage. This reduction was due to decreased amount of electrontransfer and smaller available area.In another report, the potential of the reduction peak was used an indicator of the cleanliness of1025.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical Cellsthe electrode.221 When a layer is adsorbed on the Au electrode and it acts like a barrier to electrontransfer, the potential of the reduction peak is more negative than for a bare electrode. In addition, it wasobserved that some electron transfer still occurred, due to defects in the adsorbed layer. Nevertheless,by monitoring shifts in the potential of the reduction peak, the extent of adsorbate (or contaminant)coverage of the electrode can be inferred.Electrodesorption and CoHex There are many methods available for cleaning gold electrodes forelectrochemical detection applications.218 However, the Au WE of the carboxyl-amine bonded devicescannot be directly accessed due to the enclosed PDMS cell chamber (see Figure 1.1B). For example,it would be difficult to employ the air-plasma or UV-ozone cleaning strategy. In addition, the solutionused needs to be carefully chosen, since PDMS swells in a number of solvents (see Section 1.1).Therefore, several electrochemical cleaning methods were assessed to determine their ability to removethe adsorbed TMS-EDTA layer (detailed in Section 5.3.1.7).The differential capacitance studies from Chapter 4 indicated that TMS-EDTA adsorption on Au ispotential dependent. In particular, TMS-EDTA adsorbs minimally at highly negative potentials (∼ −1.1V). Therefore, the effects of applying negative potentials to desorb TMS-EDTA were initially studied.Subsequently, a standard acid cycling cleaning method was also performed. Control experiments wereconducted using a gold bead modified with TMS-EDTA for 2 h. In addition to the two cleaning methodsdescribed, flame annealing was used as the third cleaning method for the Au bead.The reduction of hexaamminecobalt(III) (or CoHex) has been widely studied and has been found tobe irreversible.222,223 Due to its slow electron transfer kinetics,224,225 CoHex can be used as an indicatorof electrode cleanliness to test various surface cleaning strategies.The cleanliness of the Au WE inside the PDMS cell chamber was monitored with CV scans of theaqueous 750 μM CoHex solution, by carefully interrogating its reduction peak potential (EP). The highlyirreversible reduction reaction of CoHex results in a peak potential that is very sensitive to the cleanlinessof the electrode surface. When compared to a bare electrode surface, the reduction EP shifts to morenegative values in the presence of a partially blocked electrode, which reduces the electron transfer rateconstant. Figure 5.13A presents the CV scans of the gold bead electrode, and Figure 5.13B presentsthe scans of the carboxyl-amine bonded device.From Figure 5.13B, the CoHex reduction EP of the bonded device (activated and bonded using theNHS/EDC coupling chemistry) has a reduction wave at ∼ −0.3 V (solid line). After repeatedly pulsingto the negative potential in KOH, the reduction EP shifts to ∼ −0.13 V (dashed curve). This shift in the1035.3. Applying TMS-EDTA to Construct Leak-Free PDMS-Based Electrochemical CellsFigure 5.13: CoHex cyclic voltammograms of TMS-EDTA covered Au. (A) Three different methods toclean a gold bead are shown: pulsing to negative potentials (-1.4 V) in pH 13 solution, CV in 1 M sulfuricacid, and flaming. (B) Two different methods to clean the Au within a bonded PDMS device are shown:pulsing to negatives potentials in basic solution and performing CV in acidic solution. Potential scan rate= 20 mV/s.1045.4. Conclusionsreduction peak potential indicates that the electrode surface is becoming clean. Subsequent cleaningby cycling in sulfuric acid yields an electrode surface that shows a CoHex reduction EP at ∼ −0.02 V.This additional positive shift in the reduction peak potential indicates an improvement in the electrodecleanliness.For comparison, Figure 5.13A shows the CoHex CVs for a TMS-EDTA-coated gold bead electrodecleaned using the same strategies as the bonded device. In general, the Au bead shows CoHex re-duction peaks that are less negative than the bonded device. This result may be due to the lack ofNHS/EDC activation step used for PDMS bonding. Nevertheless, the reduction EP values still indicateshifts in the positive direction after the hydroxide and sulfuric acid cleaning methods. Cleaning the Aubead by flaming yields a minor change in the EP, showing that the overall cleaning procedure is quiteeffective.These results demonstrate that cleaning the bonded device using sequential basic and acidic solu-tions yields a Au surface that is comparable to the flamed bead electrode. Pressure leak tests were alsoperformed on the electrochemically cleaned devices. The leak pressures measured were comparableto the ones observed for devices without electrochemical cleaning (∼50 kPa). This result is expectedbecause electrodesorption is a displacement (or replacement) process. At the PDMS−gold interfacethat is bonded, solvent molecules (e.g., water) cannot access the gold surface. Therefore, the TMS-EDTA layer on Au still remains strongly bonded to PDMS and creates a tight seal around the PDMS cellchamber. In other words, the strong PDMS−gold bond has been retained after this Au surface cleaningmethod.5.4 ConclusionsFirst, TMS-EDTA was successfully applied to create bonded PDMS−glass and PDMS−gold structuresusing a novel carboxyl-amine bonding chemistry. Shear tests indicate that strong bonds, comparableto or better than the existing bonding techniques, can be achieved at room temperature and under mildconditions.Second, pressure leak tests were performed on the bonded 3-electrode devices, in order to obtain amore realistic measure of the bond strength under aqueous conditions. The bonding at the PDMS−Auinterface resulted in a 6-fold increase in fluidic leak pressure than air-plasma-bonded devices.Finally, a method to electrochemically clean the Au WE inside the PDMS cell chamber was devel-oped. Treatment by sequential basic and acidic solutions yields a clean Au surface that is comparable1055.4. Conclusionsto a flamed Au bead electrode.Therefore, TMS-EDTA has been successfully applied to create leak-free PDMS-based electrochem-ical cells, suitable for subsequent sensing or biosensing applications.106Chapter 6Concluding RemarksThe use of glass substrates partially patterned with a microarray of gold spots is often required formicrofluidic surface plasmon resonance (SPR) imaging applications. It is not trivial to form a strongbond between polydimethylsiloxane (PDMS) and glass while maintaining a self-assembled monolayer(SAM) on Au (inside the PDMS cell chamber) for the covalent immobilization of biomolecules. On theother hand, glass substrates decorated with sputtered Au electrodes are often used for microfluidicelectrochemical devices. During analysis, an aqueous solution of analyte(s) is injected into the PDMScell chamber. However, without treating the gold electrodes, leaks can occur at the PDMS–Au inter-face, which can severely impact the device performance. Therefore, studying substances that canself-organize onto both glass and gold surfaces is required to address some of these challenges and toenable the development of new devices for advanced biosensing and sensing applications.In this thesis, the adsorption of N-[(3-trimethoxysilyl)propyl]ethylene-diamine triacetic acid (or TMS-EDTA) on Au has been characterized and applied to construct leak-free PDMS-based electrochemicalcells. In this Chapter, a summary of the key research findings will first be recapitulated. Second, majorconclusions will be outlined. Finally, potential future work will be discussed.6.1 SummaryIn Chapter 3, the adsorption of TMS-EDTA on Au has been characterized using four complementarysurface analysis techniques: water contact angle, X-ray photoelectron spectroscopy (XPS), atomic forcemicroscopy (AFM), and infrared (IR) spectroscopy. The adsorption of a well-known alkanethiol (11-MUA)on Au has also been studied to help with the data analysis.Water contact angle measurements indicate that TMS-EDTA can chemically modify the gold surface.The resulting surface, which is similar to 11-MUA-modified Au, is more hydrophilic than bare Au (Figure3.2). Since the orientation of 11-MUA on Au is well defined (i.e., thiol interacts with Au to create terminat-ing carboxyl groups), these results suggest that TMS-EDTA may also have some terminating carboxyl1076.1. Summarygroups on Au. XPS was used to confirm the presence of carboxyl groups. The Au surfaces modifiedwith either 11-MUA or TMS-EDTA show an increase in the quantity of carbon (C) atoms in comparisonto the bare gold surface. In particular, the high resolution spectrum shows an increase in the C1s peakat 284.7 eV with an additional shoulder peak at 288 eV, which indicates the presence of carboxyl groupson TMS-EDTA-coated Au (Figure 3.3).AFM imaging was used to provide topographical and thickness information about TMS-EDTA-modifiedAu. The AFM results strongly suggest that a thin and uniform coverage of TMS-EDTA on the gold sur-face is obtained (Figure 3.4). The thickness of the TMS-EDTA layer on Au has been determined to besimilar to the 11-MUA layer on Au (Figure 3.8), which is known to be of monolayer thickness. Next, IRstudies were conducted to provide additional surface chemical information about TMS-EDTA adsorptionon Au under ambient conditions. In addition to 11-MUA, an amino-silane (3-APTMS) was also studiedto provide some insight into the extent of siloxane cross-linking (Figure 3.10). The TMS-EDTA-coatedgold surface shows a significant presence of carboxyl groups (similar to 11-MUA-coated Au) and a lackof polysiloxane formation (in contrast to 3-APTMS-coated Au).Subsequently, surface plasmon resonance (SPR) was used to provide some information regardingthe orientation of TMS-EDTA on Au. SPR experiments were conducted to demonstrate the feasibilityof using TMS-EDTA-modified Au for coupling biomolecules containing primary amino groups and to de-termine its relative stability (compared to bare Au) when a stringent wash buffer was used. The SPRresults indicate that free carboxyl groups on TMS-EDTA-modified Au are available for the immobiliza-tion of streptavidin, after carbodiimide activation (i.e., using NHS/EDC activation). Furthermore, thisunconventional surface chemistry can withstand stringent regeneration conditions (Figure 3.12).In Chapter 4, the Gibbs free energies of TMS-EDTA adsorption on Au have been quantified us-ing electrochemical methods. The TMS-EDTA-modified Au bead electrode appeared stable in DiffCapBuffer (100mMphosphate, 500mMKCl, and 65mMKOH, pH 11.6). In other words, the buffer electrolytedid not appear to significantly displace the TMS-EDTA layer over the course of about 2 hours (Figure4.3). The DiffCap Buffer was carefully formulated to ensure that the pH and the ionic concentrationsremained relatively constant during all experiments (Figure 4.5).Electrochemical differential capacitance measurements reveal that TMS-EDTA adsorption on Au ispotential-dependent (Figure 4.6). Due to pseudocapacitance features, the adsorption isotherm datawere analyzed in a narrow potential window (Figure 4.7). The Langmuir and Frumkin adsorption isothermsseem to adequately fit the data at more negative potentials. However, at more positive potentials, theFrumkin isotherm better describes the experimental data (Figure 4.8). These results are expected be-1086.1. Summarycause TMS-EDTA is negatively charged. At more positive potentials, there is more lateral interactionamong the adsorbed TMS-EDTA molecules, which is accounted for by the Frumkin adsorption isotherm.For potentials between −0.5 and 0.2 V, the potential-dependent Gibbs free energies of adsorptionwere determined to be ∼ −20 to −30 kJ/mol in the complex electrolyte solution (Figure 4.9). In thiscomplex system, adsorption is a displacement process (e.g., displacement of water molecules, as wellas phosphate, hydroxide, and chloride ions); therefore, the adsorption free energies of TMS-EDTA onAu in a non-adsorbing electrolyte are likely more favorable than the values presented here. The open-circuit potential (OCP) was measured to be ∼0.05 V and ∼ −0.025 V (vs SCE) in the presence andabsence of O2, respectively (Figure 4.1). Therefore, the free energies of adsorption of TMS-EDTA onAu around these potentials are of particular importance (since typical surface modification is done inthe presence of oxygen). These results suggest that at more positive potentials, adsorption is morethermodynamically favorable. In contrast, at highly negative potentials (∼−1.1 V), TMS-EDTA adsorbsminimally onto the Au surface.In Chapter 5, PDMS slabs were chemically modified with 3-APTMS to form terminating primaryamino groups (Figure 5.3). To demonstrate the feasibility of a novel carboxyl-amine bonding strategy,irreversible bonding was first achieved between PDMS and 11-MUA-modified Au slides (see Figure5.4). Similarly, TMS-EDTA was successfully applied to create bonded PDMS−glass and PDMS−goldstructures using the same bonding strategy. This dual functionality (i.e., achieving PDMS bonding withboth glass and gold) is not achievable by 11-MUA since it can only chemically modify Au surfaces. Sheartests indicate that strong bonds, comparable to (or stronger than) existing bonding techniques, can beachieved at room temperature and under mild conditions (Figure 5.5).Subsequently, improved surface preparation procedures were developed to fabricate robust PDMS-based electrochemical devices. TMS-EDTA was applied to chemically modify a glass substrate par-tially sputtered with Au electrodes and to create a seal around the PDMS cell chamber using thecarboxyl-amine bonding method (see Figure 5.6). Pressure leak tests were performed on the bonded3-electrode devices, in order to obtain a more realistic measure of the bond strength under aqueous con-ditions. Plasma-bonded structures were fabricated and served as controls. The results show that thePDMS−glass bond (146±6 kPa) is stronger than the PDMS−gold bond (38±9 kPa) using the carboxyl-amine strategy (see Table 5.1). For the 3-electrode substrate, the increased bonding at the PDMS−Auinterface resulted in a 6-fold increase in fluidic leak pressure (50±5 kPa) than air-plasma-bonded de-vices (8±6 kPa), which was shown to fail immediately upon loading of the aqueous sample (Figure5.12).1096.2. ConclusionsFinally, the behavior of TMS-EDTA on Au electrode at highly negative potentials (i.e., minimal adsorp-tion) was used as an advantage for preparing electrochemical sensors. After bonding the 3-electrodesubstrate with PDMS using the carboxyl-amine strategy, a method to electrochemically clean the AuWE inside the PDMS cell chamber was developed. Electrochemical cleaning using sequential basicand acidic solutions yields a clean Au surface that is comparable to the standard flamed Au bead elec-trode (Figure 5.13). Most importantly, the sealing around the PDMS cell chamber was still maintainedafter electrochemical cleaning. Therefore, the fundamental knowledge obtained about TMS-EDTA ad-sorption on Au has been successfully applied to create leak-free PDMS-based electrochemical cells,suitable for a variety of sensing and/or biosensing applications. This fabrication technology will allowthe collection of more reliable data and enable a multiple use of the device.6.2 ConclusionsDemonstrated in this thesis is the characterization of the the adsorption of a carboxylated silane (TMS-EDTA) on Au using many surface analysis techniques, resulting in a better understanding of this ad-sorption and future applications of this surface modification. The four complementary surface analysistechniques show that TMS-EDTA can be used to chemically modify the gold surface, and that the car-boxyl groups are present on Au. Furthermore, a uniform surface of monolayer-thickness is formed withlittle siloxane cross-linking. This result is similar to alkylsilanes (i.e., monolayers) but different from theamino-silane (i.e., extensively cross-linkedmultilayers) on Au (see Section 2.2.5). The presence of threenegatively charged carboxylates may contribute to this effect (i.e., the repulsion among the TMS-EDTAmolecules may prevent extensive siloxane cross-linking).The results from these initial experiments do not provide enough information about the orientationof TMS-EDTA adsorption on Au. The orientation of TMS-EDTA on Au has an effect on its utility insensor fabrication and biosensing applications. Subsequently, SPR results show that when comparedto physical immobilization (i.e., bare gold surface), TMS-EDTA plays a crucial role in capturing the largeproteins onto the Au surface using the carboxyl-amine chemistry. These results suggest that at leastsome carboxylates are oriented away from the gold surface and are available for reaction.The Gibbs free energy of TMS-EDTA adsorption on Au is also important for sensor fabrication andbiosensing applications. Subsequently, electrochemical differential capacitance was used to quantifythe potential-dependent Gibbs free energies of TMS-EDTA adsorption on Au in a complex electrolyteenvironment. The free energies of adsorption were determined to be ∼−20 to −30 kJ/mol for potentials1106.2. Conclusionsbetween −0.5 and 0.2 V, respectively. It is important to keep in mind that the adsorption of TMS-EDTAhas been interrogated with the addition of molecules and ions (e.g., water molecules and phosphate,hydroxide, and chloride ions) that also compete for the Au surface. Therefore, the TMS-EDTA adsorptionis likely more energetically favourable than the values presented here. More importantly, at more positivepotentials, the adsorption is more favorable (accompanied by increased repulsive lateral interaction). Athighly negative potentials, TMS-EDTA adsorbs minimally onto the Au electrode. This result suggeststhat the layer of TMS-EDTA may be electrochemically removed to recover a clean Au surface.It is important to note that the change in Gibbs free energy depends on the entropy change and theadsorbate/adsorbent bond strength (enthalpy change). If we assume that the entropy change of thissystem is negligible (due to the high electrolyte concentrations in the working buffer), then the apparentGibbs free energies of adsorption calculated may provide a rough estimate of the enthalpy change (i.e.,the strength of TMS-EDTA adsorption on Au). The results indicate that the typical room temperaturesurface modification of Au using aqueous solution of 10% TMS-EDTA (v/v) creates a strongly adsorbedlayer suitable for subsequent biosensor and sensor applications.The unique ability of TMS-EDTA to bond PDMS with both glass and Au surfaces has been demon-strated. Shear tests show that the bond strength of PDMS−Au structure is slightly stronger than the bondstrength of PDMS−glass structure. However, pressure leak tests show that the PDMS−glass structureproduced stronger bonds. This discrepancymay be due to the increased friction observed for shear tests(as well as the inability to observe the different modes of failure using this method). Moreover, differentsolvents were used to prepare the 3-APTMS solution (i.e., ethanol for shear tests and water for pressureleak tests) for PDMS modification. Nevertheless, since the silane-coupling chemistry for glass is wellunderstood, it is expected that TMS-EDTA reacts with glass to create a surface with three terminatingcarboxyl groups oriented away from the glass. Since the PDMS−gold bond from pressure leak tests isweaker, it is likely that the TMS-EDTA modification of Au creates a surface with some carboxylates thatare oriented away from the Au and are available for coupling to primary amino groups.Admittedly, the bond strength produced by the carboxyl-amine bonding strategy is not as strongas the air-plasma bonding method for PDMS−glass structures. However, the PDMS−gold bonding byTMS-EDTA resulted in a bond that can withstand 6X the pressure than the air-plasma strategy. Mostimportantly, the bond strength is above the maximum pressures typically encountered in microfluidic ap-plications. In the literature, bonding between PDMS and gold may be achieved when additional clean-room time is used (see Section 1.1). However, the carboxyl-amine bonding strategy is the best methodto bond a hybrid glass/gold substrate to PDMS at room temperature under ambient conditions, without1116.3. Future Workthe use of expensive equipment.Finally, due to its monolayer-thickness and little siloxane cross-linking, TMS-EDTA layer on Au WEinside the PDMS cell chamber can be remotely cleaned by applying electrochemical methods. Theproposed device fabrication protocol and the WE cleaning strategy can be used to create leak-freePDMS-based electrochemical cells, suitable for a wide range of electrochemical and/or optical sens-ing/biosensing applications.6.3 Future WorkTwo types of problems have been identified in this thesis: (1) the inability to form strong PDMS−glassbonds while forming a functional layer on the Au surface (inside the cell chamber) for surface plasmonresonance imaging (SPRi) applications, and (2) the inability to form strongPDMS−glass and PDMS−goldbonds for creating leak-free electrochemical cells (see Section 1.1). TMS-EDTA has been demonstratedto address these challenges due to its unique ability to chemically functionalize both glass and Au sur-faces. For SPRi applications, it is desirable for the TMS-EDTA layer to remain on the Au spot insidethe carboxyl-amine bonded cell chamber in order to covalently immobilize biomolecules. For electro-chemical devices, it is desirable for the TMS-EDTA layer on WE to be electrochemically removed forsubsequent sensing applications. Evidently, the strategy employed is highly dependent on the applica-tion involved. For example, other researchers have applied this novel carboxyl-amine bonding strategyusing TMS-EDTA to construct an integrated microfluidic biosensor.226 TMS-EDTA-modified Au was usedfor both PDMS bonding and covalent immobilization of antibodies for the detection of hormonal com-pounds. This result confirms the utility of TMS-EDTA in both biosensing and biosensor fabrication.Many steps of the device fabrication process can be further investigated and/or optimized. Someideas are presented below:1. Throughout this thesis, the NHS/EDC activation time and concentrations have been kept constant.The NHS/EDC activation time198 and concentrations227 may have an effect on the efficiency ofthe carboxyl-amine coupling chemistry. Therefore, more studies could be performed to determinethe optimal activation conditions for protein immobilization and bonding with 3-APTMS-modifiedPDMS.2. The effects of removing the NHS/EDC activation step may also be investigated (e.g., to determinethe extent of specific and non-specific immobilization of biomolecules).1126.3. Future Work3. In this thesis, PM-IRRAS experiments were conducted ex-situ. It would be interesting to combineelectrochemistry with PM-IRRAS experiments (i.e., polarizing the Au surface in an electrochemicalcell).4. The effect(s) of increased temperature and pressure on the bond efficiency and the bond strengthcan be tested. Furthermore, sterilization is often required for some biological studies. It would beuseful to test the bond strength after autoclaving the bonded device.5. Aqueous and ethanolic 10% 3-APTMS (v/v) solutions have been used to chemically modify PDMSsurfaces. Aqueous solution was used to reduce PDMS swelling. From the results obtained, aque-ous solution seems to produce stronger bonds for glass slides, and ethanolic solution seems toproduce stronger bonds for gold slides. Additional studies could be performed to determine theeffect(s) of different solvents on the quality of the amino-silane layer produced and the subsequentbonding strength obtained.6. The 3-electrode devices have different electrode widths (see Figure 5.10) that are bonded toPDMS. In general, it was noticed that leaks occurred more often for wider electrode widths. Thisresults suggest that electrode width has an effect on the leak pressure of a particular electrochemi-cal system. Since micrometre-scale electrodes are becoming more popular,108 solution leaks maynot be a problem for ultrathin electrode widths. Pressure leak tests could be performed on differentchannel widths to determine this effect.7. The glass and PDMS surfaces have been cleaned using UV-ozone for most of the surface modifi-cations. The use of air-plasma to clean the surface and to prepare the surface for silane function-alization may be studied.8. Currently, only a small range of potentials has been investigated for electrochemical sensing appli-cations (Figure 5.13). Additional studies could be performed to determine the effects of increasingthe potential range, which may expand the utility of this device fabrication strategy.9. After bonding the PDMS cell chambers with the 3-electrode substrate modified with TMS-EDTA,it is evident that primary amino groups are still present on the PDMS surface. The primary aminogroups make the PDMS surface less hydrophobic and more biocompatible.228,229 However, de-pending on the biosensing/sensing application pursued, the effects of primary amino groups onprobe immobilization and biomolecular interaction need to be carefully analyzed. If necessary, the1136.3. Future Workprimary amino groups may be reversibly blocked230 when the application demands a non-reactivePDMS surface.Lastly, the reaction of TMS-EDTA with other materials (e.g., silver, platinum, copper, iron, aluminum,and polymer-based materials) can be explored to further expand the application space of this carboxyl-amine bonding strategy with PDMS. The adsorption of other functional silanes on Au and other materialscan also be investigated. It is the belief of the author that the full potential of silanes has not been fullyrealized. More studies need to be performed in order to test this hypothesis.114References[1] Yang, C. W. T.; Ouellet, E.; Lagally, E. T. Anal. Chem. 2010, 82, 5408–5414.[2] Ouellet, E.; Yang, C. W. T.; Lin, T.; Yang, L. L.; Lagally, E. T. Langmuir 2010, 26, 11609–11614.[3] Yang, C. W. T.; Martens, I.; Gyenge, E. L.; Turner, R. F. B.; Bizzotto, D. J. Phys. Chem. C 2016,120, 2675–2683.[4] To, J. "Developing a Novel Heterogeneous Three ElectrodeSystem for a PDMS-BasedMicrofluidicElectrochemical Sensor." Master’s thesis, University of British Columbia, 2015 (unpublished).[5] Whitesides, G. M. Nature 2006, 442, 368–373.[6] Mark, D.; Haeberle, S.; Roth, G.; von Stetten, F.; Zengerle, R. Chem. Soc. Rev. 2010, 39, 1153–1182.[7] Puigmarti-Luis, J. Chem. Soc. Rev. 2014, 43, 2253–2271.[8] Beebe, D. J.; Mensing, G. A.; ; Walker, G. M. Annu. Rev. Biomed. Eng. 2002, 4, 261–286.[9] Paguirigan, A. L.; Beebe, D. J. Bioessays 2008, 30, 811–821.[10] Manz, A.; Harrison, D.; Verpoorte, E. M.; Fettinger, J.; Paulus, A.; Ludi, H.; Widmer, H. J. Chro-matogr. A 1992, 593, 253 – 258.[11] Nilsson, J.; Evander, M.; Hammarstrom, B.; Laurell, T. Anal. Chim. Acta 2009, 649, 141 – 157.[12] Schulte, T. H.; Bardell, R. L.; Weigl, B. H. Clin. Chim. Acta 2002, 321, 1 – 10.[13] Lion, N.; Rohner, T. C.; Dayon, L.; Arnaud, I. L.; Damoc, E.; Youhnovski, N.; Wu, Z.-Y.; Roussel, C.;Josserand, J.; Jensen, H.; Rossier, J. S.; Przybylski, M.; Girault, H. H. Electrophoresis 2003, 24,3533–3562.[14] Gulati, S.; Rouilly, V.; Niu, X.; Chappell, J.; Kitney, R. I.; Edel, J. B.; Freemont, P. S.; deMello, A. J.J. R. Soc. Interface 2009, 6, S493–S506.115References[15] Hong, J.; Edel, J. B.; deMello, A. J. Drug Discovery Today 2009, 14, 134 – 146.[16] Wang, J.-H.; Chien, L.-J.; Hsieh, T.-M.; Luo, C.-H.; Chou, W.-P.; Chen, P.-H.; Chen, P.-J.; Lee, D.-S.; Lee, G.-B. Sens. Actuators, B 2009, 141, 329 – 337.[17] Fernandes, T. G.; Diogo, M. M.; Clark, D. S.; Dordick, J. S.; Cabral, J. M. Trends Biotechnol. 2009,27, 342 – 349.[18] Ouellet, E.; Lausted, C.; Lin, T.; Yang, C. W. T.; Hood, L.; Lagally, E. T. Lab Chip 2010, 10, 581–588.[19] Thorsen, T.; Maerkl, S. J.; Quake, S. R. Science 2002, 298, 580–584.[20] Melin, J.; Quake, S. R. Annu. Rev. Biophys. Biomol. Struct. 2007, 36, 213–231.[21] Spurgeon, S. L.; Jones, R. C.; Ramakrishnan, R. PLoS One 2008, 3, 1–7.[22] Ren, K.; Zhou, J.; Wu, H. Acc. Chem. Res. 2013, 46, 2396–2406.[23] Sia, S. K.; Whitesides, G. M. Electrophoresis 2003, 24, 3563–3576.[24] McDonald, J. C.; Whitesides, G. M. Acc. Chem. Res. 2002, 35, 491–499.[25] Unger, M. A.; Chou, H.-P.; Thorsen, T.; Scherer, A.; Quake, S. R. Science 2000, 288, 113–116.[26] Deng, T.; Wu, H.; Brittain, S. T.; Whitesides, G. M. Anal. Chem. 2000, 72, 3176–3180.[27] van Kooten, T. G.; Whitesides, J. F.; von Recum, A. F. J. Biomed. Mater. Res. 1998, 43, 1–14.[28] Kim, J.-S.; Knapp, D. R. J. Am. Soc. Mass. Spectrom. 2001, 12, 463–469.[29] Thiebaud, P.; Lauer, L.; Knoll, W.; Offenhausser, A. Biosens. Bioelectron. 2002, 17, 87 – 93.[30] Paranjape, M.; Garra, J.; Brida, S.; Schneider, T.; White, R.; Currie, J. Sens. Actuators, A 2003,104, 195 – 204.[31] Yu, X.; Zhang, D.; Li, T.; Hao, L.; Li, X. Sens. Actuators, A 2003, 108, 103 – 107.[32] Mohammadi, T.; Aroujalian, A.; Bakhshi, A. Chem. Eng. Sci. 2005, 60, 1875 – 1880.[33] Abbasi, F.; Mirzadeh, H.; Katbab, A.-A. Polym. Int. 2001, 50, 1279–1287.[34] Satyanarayana, S.; Karnik, R. N.; Majumdar, A. J. Microelectromech. Syst. 2005, 14, 392–399.[35] Prats-Alfonso, E.; Albericio, F. J. Mater. Sci. 2011, 46, 7643–7648.116References[36] Robinson, T.; Verboket, P. E.; Eyer, K.; Dittrich, P. S. Lab Chip 2014, 14, 2852–2859.[37] Duffy, D. C.; McDonald, J. C.; Schueller, O. J. A.; Whitesides, G. M. Anal. Chem. 1998, 70, 4974–4984.[38] Kim, M.; Lee, S.-Y.; Choi, H.; Shin, Y.-B.; Jung, S. O.; Kim, M.-G.; Chung, B. H. Eur. Biophys. J.2006, 35, 655–662.[39] Lee, K. J.; Fosser, K. A.; Nuzzo, R. G. Adv. Funct. Mater. 2005, 15, 557–566.[40] Lim, K. S.; Chang, W.-J.; Koo, Y.-M.; Bashir, R. Lab Chip 2006, 6, 578–580.[41] Schuy, S.; Janshoff, A. J. Colloid Interface Sci. 2006, 295, 93 – 99.[42] Tang, K. C.; Liao, E.; Ong, W. L.; Wong, J. D. S.; Agarwal, A.; Nagarajan, R.; Yobas, L. J. Phys.Conf. Ser. 2006, 34, 155.[43] Zhang, Z.; Zhao, P.; Xiao, G.; Watts, B. R.; Xu, C. Biomicrofluidics 2011, 5.[44] Diaz-Gonzalez, M.; Baldi, A. Anal. Chem. 84, 7838–7844.[45] Wu, H.; Huanga, B.; Zare, R. N. Lab Chip 2005, 5, 1393–1398.[46] TermehYousefi, A.; Bagheri, S.; Adib, N. Sensor Review 2015, 35, 190–199.[47] Bang, G. S.; Cho, S.; Kim, B.-G. Biosens. Bioelectron. 2005, 21, 863 – 870.[48] Lee, J. W.; Sim, S. J.; Cho, S. M.; Lee, J. Biosens. Bioelectron. 2005, 20, 1422 – 1427.[49] Mannelli, I.; Minunni, M.; Tombelli, S.; Mascini, M. IEEE Sens. J. 2003, 3, 369–375.[50] Frey, B. L.; ; Corn, R. M. Anal. Chem. 1996, 68, 3187–3193.[51] Grabarek, Z.; Gergely, J. Anal. Biochem. 1990, 185, 131 – 135.[52] Lee, J. N.; Park, C.; Whitesides, G. M. Anal. Chem. 2003, 75, 6544–6554.[53] Dangla, R.; Gallaire, F.; Baroud, C. N. Lab Chip 2010, 10, 2972–2978.[54] Rumens, C. V.; Ziai, M. A.; Belsey, K. E.; Batchelor, J. C.; Holder, S. J. J. Mater. Chem. C 2015,3, 10091–10098.[55] Samanta, D.; Sarkar, A. Chem. Soc. Rev. 2011, 40, 2567–2592.[56] Martin, R. S.; Gawron, A. J.; Lunte, S. M.; Henry, C. S. Anal. Chem. 2000, 72, 3196–3202.117References[57] Kohler, S.; Weilbeer, C.; Howitz, S.; Becker, H.; Beushausen, V.; Belder, D. Lab Chip 2011, 11,309–314.[58] Sokolov, P.; Demidov, V.; Vedeneeva, L.; Kasyanenko, N. J. Phys. Chem. C 2015, 119, 24358–24363.[59] Vericat, C.; Vela, M. E.; Salvarezza, R. C. Phys. Chem. Chem. Phys. 2005, 7, 3258–3268.[60] Aissaoui, N.; Bergaoui, L.; Landoulsi, J.; Lambert, J.-F.; Boujday, S. Langmuir 2012, 28, 656–665.[61] Tawil, N.; Hatef, A.; Sacher, E.; Maisonneuve, M.; Gervais, T.; Mandeville, R.; Meunier, M. J. Phys.Chem. C 2013, 117, 6712–6718.[62] Addy, M.; Losey, B.; Mohseni, R.; Zlotnikov, E.; Vasiliev, A. Appl. Clay Sci. 2012, 59–60, 115–120.[63] Sharma, R.; Xu, Y.; Kim, S. W.; Schueller, M. J.; Alexoff, D.; Smith, S. D.; Wang, W.; Schlyer, D.Nanoscale 2013, 5, 7476–7483.[64] Dupont, D.; Brullot, W.; Bloemen, M.; Verbiest, T.; Binnemans, K. ACS Appl. Mater. Interfaces2014, 6, 4980–4988.[65] Dupont, D.; Luyten, J.; Bloemen, M.; Verbiest, T.; Binnemans, K. Ind. Eng. Chem. Res. 2014, 53,15222–15229.[66] Cademartiri, R.; Anany, H.; Gross, I.; Bhayani, R.; Griffiths, M.; Brook, M. A. Biomaterials 2010,31, 1904–1910.[67] Spencer, M. J.; Nyberg, G. L. Surf. Sci. 2004, 573, 151 – 168.[68] Yochelis, S.; Katzir, E.; Kalcheim, Y.; Gutkin, V.; Millo, O.; Paltiel, Y. J. Nanotechnol. 2012, 2012,8.[69] Chen, Y.; Holl, M. M. B.; Orr, B. G. Surf. Sci. 2007, 601, 1937 – 1943.[70] Bahadira, E. B.; Sezginturk, M. K. Anal. Biochem. 2015, 478, 107 – 120.[71] Yanez-Sedeno, P.; Agui, L.; Villalonga, R.; Pingarron, J. Anal. Chim. Acta 2014, 823, 1 – 19.[72] Roda, A.; Michelini, E.; Zangheri, M.; Fusco, M. D.; Calabria, D.; Simoni, P. TrAC, Trends Anal.Chem. 2016, 79, 317 – 325, Past , Present and Future challenges of Biosensors and Bioanalyticaltools in Analytical Chemistry: a tribute to Prof Marco Mascini.118References[73] Rotariu, L.; Lagarde, F.; Jaffrezic-Renault, N.; Bala, C. TrAC, Trends Anal. Chem. 2016, 79, 80– 87, Past , Present and Future challenges of Biosensors and Bioanalytical tools in AnalyticalChemistry: a tribute to Prof Marco Mascini.[74] Sin, M. L.; Mach, K. E.; Wong, P. K.; Liao, J. C. Expert Rev. Mol. Diagn. 2014, 14, 225–244.[75] Zhang, J.; Jensen, M. K.; Keasling, J. D. Curr. Opin. Chem. Biol. 2015, 28, 1 – 8.[76] Goode, J. A.; Rushworth, J. V. H.; Millner, P. A. Langmuir 2015, 31, 6267–6276.[77] Zhang, D.; Liu, Q. Biosens. Bioelectron. 2016, 75, 273 – 284.[78] Bahadira, E. B.; Sezginturk, M. K. Biosens. Bioelectron. 2015, 68, 62 – 71.[79] Yoo, S. M.; Lee, S. Y. Trends Biotechnol. 2016, 34, 7 – 25.[80] Chao, J.; Zhu, D.; Zhang, Y.; Wang, L.; Fan, C. Biosens. Bioelectron. 2016, 76, 68 – 79.[81] Justino, C. I.; Freitas, A. C.; Pereira, R.; Duarte, A. C.; Santos, T. A. R. TrAC, Trends Anal. Chem.2015, 68, 2 – 17.[82] Liu, Q.; Wang, J.; Boyd, B. J. Talanta 2015, 136, 114 – 127.[83] Mehrotra, P. J. Oral Biol. Craniofac. Res. 2016, 6, 153 – 159.[84] Acquah, C.; Danquah, M. K.; Yon, J. L.; Sidhu, A.; Ongkudon, C. M. Anal. Chim. Acta 2015, 888,10 – 18.[85] Coad, B. R.; Jasieniak, M.; Griesser, S. S.; Griesser, H. J. Surf. Coat. Technol. 2013, 233, 169 –177.[86] Huang, J.; Yang, X.; He, X.; Wang, K.; Liu, J.; Shi, H.; Wang, Q.; Guo, Q.; He, D. TrAC, TrendsAnal. Chem. 2014, 53, 11 – 20.[87] Jia, F.; Narasimhan, B.; Mallapragada, S. Biotechnol. Bioeng. 2014, 111, 209–222.[88] Makaraviciute, A.; Ramanaviciene, A. Biosens. Bioelectron. 2013, 50, 460 – 471.[89] Meyer, R.; Giselbrecht, S.; Rapp, B. E.; Hirtz, M.; Niemeyer, C. M. Curr. Opin. Chem. Biol. 2014,18, 8 – 15.[90] Desmet, C.; Marquette, C. A.; Blum, L. J.; Doumeche, B. Biosens. Bioelectron. 2016, 76, 145 –163.119References[91] Teles, F.; Fonseca, L. Mater. Sci. Eng., C 2008, 28, 1530 – 1543.[92] Zhou, Y.; Chiu, C.-W.; Liang, H. Sensors 2012, 12, 15036.[93] Nakanishi, K.; Sakiyama, T.; Kumada, Y.; Imamura, K.; Imanaka, H. Curr. Proteomics 2008, 5,161–175.[94] Banuls, M.-J.; Puchades, R.; Maquieira, A. Anal. Chim. Acta 2013, 777, 1 – 16.[95] Wijaya, E.; Lenaerts, C.; Maricot, S.; Hastanin, J.; Habraken, S.; Vilcot, J.-P.; Boukherroub, R.;Szunerits, S. Curr. Opin. Solid State Mater. Sci. 2011, 15, 208 – 224.[96] Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105,1103–1170.[97] Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775–1789.[98] Sassolas, A.; Blum, L. J.; Leca-Bouvier, B. D. Biotechnol. Adv. 2012, 30, 489 – 511.[99] Patel, N.; Davies, M. C.; Hartshorne, M.; Heaton, R. J.; Roberts, C. J.; Tendler, S. J. B.;Williams, P. M. Langmuir 1997, 13, 6485–6490.[100] Vericat, C.; Vela, M. E.; Benitez, G.; Carro, P.; Salvarezza, R. C. Chem. Soc. Rev. 2010, 39,1805–1834.[101] Lane, T. H.; Burns, S. A. In Immunology of Silicones; Potter, M., Rose, N. R., Eds.; SpringerBerlin Heidelberg: Berlin, Heidelberg, 1996; Chapter Silica, Silicon and Silicones...Unraveling theMystery, pp 3–12.[102] Kregiel, D. Food Control 2014, 40, 32 – 40.[103] Saner, C. K.; Lusker, K. L.; LeJeune, Z. M.; Serem, W. K.; Garno, J. C. Beilstein J. Nanotechnol.2012, 3, 114–122.[104] Helmy, R.; Wenslow, R. W.; Fadeev, A. Y. JACS 2004, 126, 7595–7600.[105] Dreyfuss, P. Macromolecules 1978, 11, 1031–1036.[106] Dreyfuss, P.; Fetters, L. J.; Gent, A. N. Macromolecules 1978, 11, 1036–1038.[107] Bel-Hassen, R.; Boufi, S.; Salon, M.-C. B.; Abdelmouleh, M.; Belgacem, M. N. J. Appl. Polym.Sci. 2008, 108, 1958–1968.120References[108] Rackus, D. G.; Shamsi, M. H.; Wheeler, A. R. Chem. Soc. Rev. 2015, 44, 5320–5340.[109] Seethapathy, S.; Gorecki, T. Anal. Chim. Acta 2012, 750, 48 – 62.[110] McDonald, J. C.; Duffy, D. C.; Anderson, J. R.; Chiu, D. T.; Wu, H.; Schueller, O. J. A.; White-sides, G. M. Electrophoresis 2000, 21, 27–40.[111] Xiong, L.; Chen, P.; Zhou, Q. J. Adhes. Sci. Technol. 2014, 28, 1046–1054.[112] Zhou, J.; Khodakov, D. A.; Ellis, A. V.; Voelcker, N. H. Electrophoresis 2012, 33, 89–104.[113] Vig, J. R. J. Vac. Sci. Technol., A 1985, 3, 1027–1034.[114] Zhou, J.; Ellis, A. V.; Voelcker, N. H. Electrophoresis 2010, 31, 2–16.[115] Phillips, K. S.; Cheng, Q. Anal. Bioanal.Chem. 2007, 387, 1831–1840.[116] Green, R. J.; Frazier, R. A.; Shakesheff, K. M.; Davies, M. C.; Roberts, C. J.; Tendler, S. J. Bio-materials 2000, 21, 1823 – 1835.[117] Homola, J. Chem. Rev. 2008, 108, 462–493.[118] Cooper, M. A. Nat. Rev. Drug Discov. 2002, 1, 515–528.[119] Katz, C.; Levy-Beladev, L.; Rotem-Bamberger, S.; Rito, T.; Rudiger, S. G. D.; Friedler, A. Chem.Soc. Rev. 2011, 40, 2131–2145.[120] Wain, A. J.; Do, H. N. L.; Mandal, H. S.; Kraatz, H.-B.; Zhou, F. J. Phys. Chem. C 2008, 112,14513–14519.[121] Fischer, M. J. E. In Surface Plasmon Resonance: Methods and Protocols; Mol, J. N., Fischer, E.M. J., Eds.; Humana Press: Totowa, NJ, 2010; Chapter Amine Coupling Through EDC/NHS: APractical Approach, pp 55–73.[122] Galopin, E.; Beaugeois, M.; Pinchemel, B.; Camart, J.-C.; Bouazaoui, M.; Thomy, V. Biosens.Bioelectron. 2007, 23, 746 – 750.[123] Choi, S. H.; Lee, J. W.; Sim, S. J. Biosens. Bioelectron. 2005, 21, 378 – 383.[124] Real-Fernandez, F.; Rossi, G.; Lolli, F.; Papini, A. M.; Rovero, P. MethodsX 2015, 2, 141 – 144.[125] Raz, S. R.; Bremer, M. G.; Giesbers, M.; Norde, W. Biosens. Bioelectron. 2008, 24, 552 – 557.121References[126] Chinowsky, T. M.; Grow, M. S.; Johnston, K. S.; Nelson, K.; Edwards, T.; Fu, E.; Yager, P. Biosens.Bioelectron. 2007, 22, 2208 – 2215.[127] Luo, Y.; Yu, F.; Zare, R. N. Lab Chip 2008, 8, 694–700.[128] Linman, M. J.; Abbas, A.; Cheng, Q. Analyst 2010, 135, 2759–2767.[129] Krishnamoorthy, G.; Carlen, E. T.; Bomer, J. G.; Wijnperle, D.; deBoer, H. L.; van den Berg, A.;Schasfoort, R. B. M. Lab Chip 2010, 10, 986–990.[130] Rich, R. L.; Cannon, M. J.; Jenkins, J.; Pandian, P.; Sundaram, S.; Magyar, R.; Brockman, J.;Lambert, J.; Myszka, D. G. Anal. Biochem. 2008, 373, 112 – 120.[131] Piliarik, M.; Vaisocherov, H.; Homola, J. Biosens. Bioelectron. 2005, 20, 2104–2110.[132] Krishnamoorthy, G.; Bianca Beusink, J.; Schasfoort, R. B. M. Anal. Methods 2010, 2, 1020–1025.[133] Somorjai, G. A.; Li, Y. Introduction to surface chemistry and catalysis; John Wiley & Sons, 2010.[134] Yuan, Y.; Lee, T. R. Surface science techniques; Springer, 2013; pp 3–34.[135] Gupta, P.; Ulman, A.; Fanfan, S.; Korniakov, A.; Loos, K. JACS 2005, 127, 4–5.[136] Bodas, D.; Khan-Malek, C. Sens. Actuators, B 2007, 123, 368 – 373.[137] Giessibl, F. J. Rev. Mod. Phys. 2003, 75, 949–983.[138] Geisse, N. A. Mater. Today 2009, 12, 40 – 45.[139] Morris, V.; Kirby, A.; Gunning, A. 2010,[140] Sasaki, K.; Koike, Y.; Azehara, H.; Hokari, H.; Fujihira, M. Appl. Phys. A 1998, 66, 1275–1277.[141] Shahin, V.; Ludwig, Y.; Schafer, C.; Nikova, D.; Oberleithner, H. J. Cell Sci. 2005, 118, 2881–2889.[142] Hu, K.; Bard, A. J. Langmuir 1997, 13, 5114–5119.[143] Stuart, B. Kirk-Othmer Encyclopedia of Chemical Technology; John Wiley & Sons, Inc., 2000.[144] Ramin, M. A.; Bourdon, G. L.; Daugey, N.; Bennetau, B.; Vellutini, L.; Buffeteau, T. Langmuir2011, 27, 6076–6084.[145] Hind, A. R.; Bhargava, S. K.; McKinnon, A. Adv. Colloid Interface Sci. 2001, 93, 91–114.[146] Mendelsohn, R.; Mao, G.; Flach, C. R. Biochim. Biophys. Acta 2010, 1798, 788–800.122References[147] Golden, W. G.; Saperstein, D. D.; Severson, M. W.; Overend, J. J. Phys. Chem. 1984, 88, 574–580.[148] Chusuei, C. C.; Goodman, D. W. Encyclopedia of physical science and technology 2002, 17,921–938.[149] Finklea, H. O.; Robinson, L. R.; Blackburn, A.; Richter, B.; Allara, D.; Bright, T. Langmuir 1986, 2,239–244.[150] Owens, T. M.; Nicholson, K. T.; Fosnacht, D. R.; Orr, B. G.; Holl, M. M. B. Langmuir 2006, 22,9619–9622.[151] Arkles, B.; Pan, Y.; Mi Kim, Y.; Eisenbraun, E.; Miller, C.; Kaloyeros, A. E. J. Adhes. Sci. Technol.2012, 26, 41–54.[152] Kurth, D. G.; Bein, T. Langmuir 1995, 11, 3061–3067.[153] Owens, T. M.; Nicholson, K. T.; Holl, M. M. B.; Suzer, S. JACS 2002, 124, 6800–6801.[154] Owens, T. M.; Suzer, S.; Holl, M. M. B. J. Phys. Chem. B 2003, 107, 3177–3182.[155] Owens, T. M.; Ludwig, B. J.; Schneider, K. S.; Fosnacht, D. R.; Orr, B. G.; Holl, M. M. B. Langmuir2004, 20, 9636–9645.[156] Barnes, G.; Gentle, I. Interfacial science: an introduction; Oxford University Press, 2011.[157] Wang, F.; Stuart, S. J.; Latour, R. A. Biointerphases 2008, 3, 9–18.[158] Thyparambil, A. A.; Wei, Y.; Latour, R. A. Langmuir 2012, 28, 5687–5694.[159] Liu, Y. J. Chem. Eng. Data 2009, 54, 1981–1985.[160] Ferrari, L.; Kaufmann, J.; Winnefeld, F.; Plank, J. J. Colloid Interface Sci. 2010, 347, 15 – 24.[161] Ghosh, P. Colloid and Interface Science; PHI Learning, 2009.[162] Srinivasan, R.; Suni, I. J. Appl. Electrochem. 1998, 28, 993–998.[163] Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Principles and Applications 2001, 386–428.[164] Stern, O. Z. Elektrochem 1924, 30, 1014–1020.[165] Grahame, D. C. Chem. Rev. 1947, 41, 441–501.123References[166] Bockris, J. O.; Devanathan, M. A. V.; Muller, K. Proc. R. Soc. Lond. A Math. Phys. Sci. 1963, 274,55–79.[167] Wu, J.; Risalvato, F. G.; Ke, F.-S.; Pellechia, P. J.; Zhou, X.-D. J. Electrochem. Soc. 2012, 159,F353–F359.[168] Mohilner, D. M. J. Phys. Chem. 1962, 66, 724–726.[169] Richer, J.; Stolberg, L.; Lipkowski, J. Langmuir 1986, 2, 630–638.[170] Shi, Z.; Lipkowski, J. J. Electroanal. Chem. 1996, 403, 225.[171] Chen, A.; ; Lipkowski, J. J. Phys. Chem. B 1999, 103, 682–691.[172] Conway, B. E.; Pell, W. G. J. Solid State Electrochem. 2003, 7, 637–644.[173] Andrieux, C. P.; Hapiot, P.; Saveant, J. M. Chem. Rev. 1990, 90, 723–738.[174] Taylor, A. W.; Qiu, F.; Hu, J.; Licence, P.; Walsh, D. A. J. Phys. Chem. B 2008, 112, 13292–13299.[175] Dowsett, A. A. . M. Acc. Chem. Res. 2010, 43, 927–935.[176] Xiao, W.; Wang, D. Chem. Soc. Rev. 2014, 43, 3215–3228.[177] Xiang, Y.; Lu, S.; Jiang, S. P. Chem. Soc. Rev. 2012, 41, 7291–7321.[178] Mayer, M. T.; Lin, Y.; Yuan, G.; Wang, D. Acc. Chem. Res. 2013, 46, 1558–1566.[179] Arias, Z. G.; Alvarez, J. L. M.; Fonseca, J. M. L. Electroanalysis 2004, 16, 1044–1050.[180] Choi, K.; Ng, A. H.; Fobel, R.; Wheeler, A. R. Annu. Rev. Anal. Chem. 2012, 5, 413–440.[181] Li, M.; Li, Y.-T.; Li, D.-W.; Long, Y.-T. Anal. Chim. Acta 2012, 734, 31 – 44.[182] Tudorache, M.; Bala, C. Anal. Bioanal.Chem. 2007, 388, 565–578.[183] Heller, A.; Feldman, B. Chem. Rev. 2008, 108, 2482–2505.[184] Wang, J.; Musameh, M. Anal. Lett. 2003, 36, 2041–2048.[185] Dydek, E. V.; Petersen, M. V.; Nocera, D. G.; Jensen, K. F. J. Electrochem. Soc. 2012, 159,H853–H856.[186] Pumera, M.; Merkoci, A.; Alegret, S. TrAC, Trends Anal. Chem. 2006, 25, 219 – 235.124References[187] Kim, S. K.; Lim, H.; Chung, T. D.; Kim, H. C. Sens. Actuators, B 2006, 115, 212 – 219.[188] Kang, C. M.; Joo, S.; Bae, J. H.; Kim, Y.-R.; Kim, Y.; Chung, T. D. Anal. Chem. 2012, 84, 901–907.[189] Stalder, A. F.; Melchior, T.; Muller, M.; Sage, D.; Blu, T.; Unser, M. Colloids Surf., A 2010, 364, 72– 81.[190] Stettner, J.; Frank, P.; Griesser, T.; Trimmel, G.; Schennach, R.; Resel, R.; Winkler, A. In InterfaceControlled Organic Thin Films; Al-Shamery, K., Horowitz, G., Sitter, H., Rubahn, H.-G., Eds.;Springer Berlin Heidelberg: Berlin, Heidelberg, 2009; Chapter Characterization of 11-MUA SAMformation on gold surfaces, pp 101–105.[191] Mendoza, S. M.; Arfaoui, I.; Zanarini, S.; Paolucci, F.; Rudolf, P. Langmuir 2007, 23, 582–588.[192] Marciniec, B.; Krysztafkiewicz, A.; Domka, L. Colloid. Polym. Sci. 1983, 261, 306–311.[193] Haensch, C.; Hoeppener, S.; Schubert, U. S. Chem. Soc. Rev. 2010, 39, 2323–2334.[194] Sterman, S.; Bradley, H. B. Polym. Eng. Sci. 1961, 1, 224–233.[195] Cecchet, F.; Rudolf, P.; Rapino, S.; Margotti, M.; Paolucci, F.; Baggerman, J.; Brouwer, A. M.;Kay, E. R.; Wong, J. K. Y.; Leigh, D. A. J. Phys. Chem. B 2004, 108, 15192–15199.[196] Sam, S.; Touahir, L.; Andresa, J. S.; Allongue, P.; Chazalviel, J. N.; Gouget-Laemmel, A.; de Vil-leneuve Henry,; Moraillon, A.; Ozanam, F.; Gabouze, N.; Djebbar, S. Langmuir 2010, 26, 809–814.[197] de Oliveira, R. M.; Ferreira, J.; Santos, M. J. L.; Faria, R. M.; Oliveira, O. N. ChemPhysChem2011, 12, 1736–1740.[198] Palazon, F.; Benavides, C. M.; Leonard, D.; Souteyrand, E.; Chevolot, Y.; Cloarec, J.-P. Langmuir2014, 30, 4545–4550.[199] Simsikova, M.; Antalik, M.; Kanuchova, M.; Skvarla, J. Appl. Surf. Sci. 2013, 282, 342 – 347.[200] Uluok, S.; Guven, B.; Eksi, H.; Ustundag, Z.; Tamer, U.; Boyaci, I. H. J. Nanopart. Res. 2015, 17,1–12.[201] Bhadra, P.; Shajahan, M. S.; Bhattacharya, E.; Chadha, A. RSC Adv. 2015, 5, 80480–80487.[202] Jin, Y.; Kang, X.; Song, Y.; Zhang, B.; Cheng, G.; Dong, S. Anal. Chem. 2001, 73, 2843–2849.125References[203] Du, K.; Wathuthanthri, I.; Liu, Y.; Xu, W.; Choi, C.-H. ACS Appl. Mater. Interfaces 2012, 4, 5505–5514.[204] Katsonis, N.; Marchenko, A.; Taillemite, S.; Fichou, D.; Chouraqui, G.; Aubert, C.; Malacria, M.Chemistry 2003, 9, 2574–2581.[205] Shepherd, J.; Yang, Y.; Bizzotto, D. J. Electroanal. Chem. 2002, 524–525, 54–61.[206] Lemay, D. M.; Shepherd, J. L. Electrochim. Acta 2008, 54, 388–393.[207] Markin, V. S.; Volkova-Gugeshashvili, M.; Volkov, A. G. J. Phys. Chem. B 2006, 110, 11415–11420.[208] Yoo, H. D.; Jang, J. H.; Ka, B. H.; Rhee, C. K.; Oh, S. M. Langmuir 2009, 25, 11947–11954.[209] Chen, H.-Y.; McClelland, A. A.; Chen, Z.; Lahann, J. Anal. Chem. 2008, 80, 4119–4124.[210] Eddings, M. A.; Johnson, M. A.; Gale, B. K. J. Micromech. Microeng. 2008, 18, 067001.[211] Vilaplana, F.; Osorio-Galindo, M.; Iborra-Clar, A.; Alcaina-Miranda, M. I.; Ribes-Greus, A. J. Appl.Polym. Sci. 2004, 93, 1384–1393.[212] Lee, N. Y.; Chung, B. H. Langmuir 2009, 25, 3861–3866.[213] Cabral, J.; Kennedy, J. Bioprocess Technol. 1991, 14, 73.[214] Shechter, L.; Wynstra, J. Ind. Eng. Chem. 1956, 48, 86–93.[215] Ng, S. H.; Tjeung, R. T.; Wang, Z. F.; Lu, A. C. W.; Rodriguez, I.; Rooij, N. F. Microsyst. Technol.2007, 14, 753–759.[216] Hohertz, D.; Romanuik, S. F.; Gray, B. L.; Kavanagh, K. L. J. Vac. Sci. Technol., A 2014, 32.[217] Bhattacharya, S.; Datta, A.; Berg, J. M.; Gangopadhyay, S. J. Microelectromech. Syst. 2005, 14,590–597.[218] Fischer, L. M.; Tenje, M.; Heiskanen, A. R.; Masuda, N.; Castillo, J.; Bentien, A.; Emneusa, J.;Jakobsen, M. H.; Boisen, A. Microelectron. Eng. 2009, 86, 1282 – 1285.[219] Gueshi, T.; Tokuda, K.; Matsuda, H. J. Electroanal. Chem. Interfacial Electrochem. 1979, 101, 29– 38.126[220] Amatore, C.; Saveant, J.; Tessier, D. J. Electroanal. Chem. Interfacial Electrochem. 1983, 147, 39– 51.[221] Bizzotto, D.; McAlees, A.; Lipkowski, J.; McCrindle, R. Langmuir 1995, 11, 3243–3250.[222] Ogura, K.; Fukushima, Y. Electrochim. Acta 1981, 26, 93 – 97.[223] Hamelin, A.; Weaver, M. J. J. Electroanal. Chem. Interfacial Electrochem. 1986, 209, 109 – 124.[224] Hromadova, M.; Fawcett, W. R. J. Phys. Chem. A 2000, 104, 4356–4363.[225] Ji, X.; Chevallier, F. G.; Clegg, A. D.; Buzzeo, M. C.; Compton, R. G. J. Electroanal. Chem. 2005,581, 249 – 257.[226] Pires, N. M. M.; Dong, T.; Hanke, U.; Hoivik, N. J. Biomed. Opt. 2013, 18, 097001–097001.[227] Booth, M. A.; Kannappan, K.; Hosseini, A.; Partridge, A. Langmuir 2015, 31, 8033–8041.[228] Kuddannaya, S.; Chuah, Y. J.; Lee, M. H. A.; Menon, N. V.; Kang, Y.; Zhang, Y. ACS Appl. Mater.Interfaces 2013, 5, 9777–9784.[229] Kuddannaya, S.; Bao, J.; Zhang, Y. ACS Appl. Mater. Interfaces 2015, 7, 25529–25538.[230] Butler, P. J. G.; Harris, J. I.; Hartley, B. S.; Leberman, R. Biochem. J 1969, 112, 679–689.[231] Smith, C. P.; White, H. S. Langmuir 1993, 9, 1–3.[232] Burgess, I.; Seivewright, B.; Lennox, R. B. Langmuir 2006, 22, 4420–4428.127Appendix AAppendix for Chapter 3The aqueous 10% TMS-EDTA (v/v) solution is basic as indicated by Figure A.1 (prior to the addition of1M HCl). The titration curve for adding 1MHCl to 10% TMS-EDTA (v/v) solution is shown. The derivative(ΔpHΔV) is plotted against average volume. The pK values of oxide-based nanoparticles modified withTMS-EDTA have been determined to be 4.17, 6.89, and 10.00.64 It is known that the pK values aredifferent for solution-based and surface-adsorbed 11-MUA layer.231,232 Therefore, it is not surprisingthat the pK values are somewhat different. Nevertheless, at highly basic pH values (i.e., 11.6), theTMS-EDTA molecules are mostly deprotonated (i.e., negatively charged).-12-10-8-6-4-200246810120 1 2 3 4 5 6 7 8 9 10pHVolume of HCl Added (mL)Titration Curve of TMS-EDTApHP‰,lPsFigure A.1: Titration curve for adding 1M HCl to 10% TMS-EDTA (v/v) solution.128Appendix BAppendix for Chapter 4To ensure high reproducibility and cleanliness of the three-electrode electrochemical setup, full range CVof the DiffCap buffer (see Figure B.1) was obtained before the addition of any TMS-EDTA. Subsequently,double layer CV was obtained before and after the addition of bulk TMS-EDTA concentrations for allexperiments (see Figure B.2 for two examples).Once the system has been determined to be clean, electrochemical differential capacitance mea-surements were conducted. A total of 19 TMS-EDTA concentrations were investigated: Buffer (0 μM),1 μM, 2 μM, 5 μM, 10 μM, 20 μM, 50 μM, 100 μM, 200 μM, 500 μM, 1 mM, 2 mM, 5 mM, 10 mM, 20mM, 50 mM, 100 mM, 150 mM, and 200 mM. At each TMS-EDTA concentration, a total of 19 valuesof potential of interest (POI) were examined: −1.1, −1.0, −0.9, −0.8, −0.7, −0.6, −0.5, −0.45, −0.4,−0.35, −0.3, −0.25, −0.2, −0.15, −0.10, −0.05, 0, 0.1, and 0.2 V.A portion of the algorithm used to collect the data (using NOVA 1.8 software) at a particular POI isshown in Table B.1. This algorithm was repeated for all 19 potentials from −1.1 to 0.2 V for a constantTMS-EDTA concentration. The potential-dependent capacitance measured at a particular TMS-EDTAconcentration took about 2 hours and 40 minutes to complete. This process was repeated for all 19concentrations, with triplicates for lower TMS-EDTA concentrations.The raw differential capacitance data for DiffCap Buffer and some representative TMS-EDTA con-centrations are shown in Figure B.3 to Figure B.9. Briefly, capacitance was measured for a range of 19potentials (starting from−1.1 V and stepping positively to 0.2 V). At each potential, stirring was turnedon for 60 s and turned off prior to measuring the capacitance (repeated six times). After 6 min of stirring,the potential-dependent adsorption of TMS-EDTA on Au was assumed to have reached an equilibrium(as determined by the 200 mM TMS-EDTA raw differential capacitance data).The area of the Au bead electrode was estimated using the method based on Shepherd et al.’sstudy:2051. Capacitance curves of 50 mM Perchlorate Buffer (pH 12) and DiffCap Buffer (pH 12) were obtained(see Figure B.10).129Appendix B. Appendix for Chapter 4Figure B.1: Full range CV of the DiffCap buffer was obtained.Table B.1: The algorithm used to collect the electrochemical differential capacitance measurements(using NOVA 1.8 software) at one potential of interest (POI).Potential (V) Time (s) Stirrer Notes−1.1 30 Off Desorption to ensure same initial conditionsPOI 60 On Adsorption with stirring at POI (1 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)POI 60 On Adsorption with stirring at POI (2 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)POI 60 On Adsorption with stirring at POI (3 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)POI 60 On Adsorption with stirring at POI (4 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)POI 60 On Adsorption with stirring at POI (5 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)POI 60 On Adsorption with stirring at POI (6 min total)POI 10 Off Allowing system to reach equilibriumPOI 10 Off Measuring 200 data points (at 0.05 s intervals)130Appendix B. Appendix for Chapter 4Figure B.2: Double layer CV of the DiffCap buffer was obtained before and after the addition of bulkTMS− EDTA concentrations for all experiments. Two TMS− EDTA concentrations (20 μM and 20mM) are shown.Figure B.3: Raw differential capacitance data of DiffCap buffer.131Appendix B. Appendix for Chapter 4Figure B.4: Raw differential capacitance data of 2 μM TMS-EDTA.Figure B.5: Raw differential capacitance data of 20 μM TMS-EDTA.132Appendix B. Appendix for Chapter 4Figure B.6: Raw differential capacitance data of 200 μM TMS-EDTA.Figure B.7: Raw differential capacitance data of 2 mM TMS-EDTA.133Appendix B. Appendix for Chapter 4Figure B.8: Raw differential capacitance data of 20 mM TMS-EDTA.Figure B.9: Raw differential capacitance data of 200 mM TMS-EDTA.134Appendix B. Appendix for Chapter 4Figure B.10: Capacitance curves of 50 mM Perchlorate Buffer (solid line) and DiffCap Buffer (dottedline) at pH 12 are shown. These curves were used to estimate the area of the Au bead in DiffCap Buffer.2. The raw capacitance value of Perchlorate Buffer at −0.9 V was determined to be ∼4.26 μF.3. The capacitance value from Step (2) was divided by by 17 μF/cm2 (a value proposed to be con-stant for the negative potentials of Perchlorate Buffer) in order to obtain an estimate of the area ofthe Au bead electrode.4. An electrode area of ∼0.25 cm2 was determined.The adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line) andFrumkin (solid line) isotherms, for applied electrode potentials between −1.1 and −0.6 are shown inFigures B.11 to Figure B.16. Although the data fitting converged to a numerical value for these potentials,the surface coverage data shown Figure 4.7A reveal that pseudocapacitance features adversely affectthe data analysis. Therefore, the analysis was restricted to a range of potentials between −0.5 to 0.2 V.Finally, the R2 values from the Langmuir and Frumkin fittings are shown in Figures B.17 and FigureB.18.135Appendix B. Appendix for Chapter 410−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −1.1 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.11: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −1.1 V.10−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −1 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.12: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −1.0 V.136Appendix B. Appendix for Chapter 410−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −0.9 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.13: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −0.9 V.10−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −0.8 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.14: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −0.8 V.137Appendix B. Appendix for Chapter 410−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −0.7 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.15: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −0.7 V.10−6 10−5 10−4 10−3 10−2 10−100.20.40.60.81Langmuir and Frumkin Fitted Curves for: −0.6 VConcentration (M)Fractional Surface Coverage θ  Experimental DataLangmuir FittingFrumkin FittingFigure B.16: Adsorption of TMS-EDTA onto Au electrode in DiffCap Buffer, fit with Langmuir (dotted line)and Frumkin (solid line) isotherms, at applied electrode potentials of −0.6 V.138Appendix B. Appendix for Chapter 4Figure B.17: The potential-dependent R2 values from the Langmuir non-linear least squares fitting.Figure B.18: The potential-dependent R2 values from the Frumkin non-linear least squares fitting.139Appendix CAppendix for Chapter 5Three elastic failure curves (from three different samples) were obtained for each of the bonding methodusing shear tests, and standard deviation was calculated using these values. The force values were thenconverted to pressures by dividing the PDMS−substrate contact area. Table C.1 shows the results fromshear tests of glass−PDMS bonding strategies.Table C.1: Results from shear tests of glass−PDMS bonding strategies.Glass Substrate Modification PDMS Modification Force Recorded (N) Bond Strength (kPa)UV-Ozone UV-Ozone 87.4±10 271.2±30.9Oxygen Plasma Oxygen Plasma 149.1±12.5 462.2±38.3TMS-EDTA 3-APTMS 92.5±5.6 286.6±17.2BTMSE UV-Ozone 163.5±21.9 506.9±67.9Table C.2 shows the results from shear tests of gold−PDMS bonding strategies.Table C.2: Results from shear tests of gold−PDMS bonding strategies.Gold Substrate Modification PDMS Modification Force Recorded (N) Bond Strength (kPa)3-MPTMS UV-Ozone 122.7±21.1 280.2±65.411-MUA 3-APTMS 103.1±14.6 319.7±45.3TMS-EDTA 3-APTMS 104.9±6.3 325.3±19.4Additional control experiments were conducted by manual peel tests to determine the importance offorming both terminal carboxyl groups and terminal primary amino groups on solid substrate (i.e., glassor gold) and PDMS, respectively. Table C.3 shows the results from these manual peel tests.A schematic of the setup for shear tests is shown in Figure C.1. Since it was difficult to determine thedifferent modes of failure from shear test results, pressure leak tests were conducted on devices withPDMS cell chambers bonded to 3-electrode substrates to obtain a more realistic measure of the bondstrength under aqueous conditions. An example of the pressure leak test experiment is shown in FigureC.2. Each of the three electrodes failed at different pressures. The sharp drop in pressure near the endindicated that the solution had leaked beyond the PDMS−Au interface.140Appendix C. Appendix for Chapter 5Table C.3: Results from manual peel tests of gold−PDMS bonding strategies.Gold Substrate Modification PDMS Modification ResultUV-Ozone UV-Ozone No bonding observedUV-Ozone 3-APTMS No bonding observed11-MUA UV-Ozone No bonding observedTMS-EDTA UV-Ozone No bonding observedUV-Ozone TMS-EDTA No bonding observedTMS-EDTA 3-APTMS Irreversible bondingBTMSE UV-Ozone Irreversible bonding3-MPTMS UV-Ozone Irreversible bondingUV-Ozone 3-MPTMS Irreversible bonding141Appendix C. Appendix for Chapter 5Figure C.1: (A) A schematic of a typical sample used for shear tests: a PDMS slab bonded to a glasssubstrate is shown. (B) Top view of the PDMS-bonded glass substrate. (C) A side-view diagram of theshear test procedure.142Appendix C. Appendix for Chapter 5Figure C.2: An example of pressure leak test experiment for carboxyl-amine bonded PDMS-based 3-electrode device.143

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0308091/manifest

Comment

Related Items