Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electrochemically controlled interaction of liposomes with a solid-supported octadecanol bilayer Musgrove, Amanda 2013

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

Item Metadata


24-ubc_2013_fall_musgrove_amanda.pdf [ 9.87MB ]
JSON: 24-1.0165606.json
JSON-LD: 24-1.0165606-ld.json
RDF/XML (Pretty): 24-1.0165606-rdf.xml
RDF/JSON: 24-1.0165606-rdf.json
Turtle: 24-1.0165606-turtle.txt
N-Triples: 24-1.0165606-rdf-ntriples.txt
Original Record: 24-1.0165606-source.json
Full Text

Full Text

Electrochemically Controlled Interaction ofLiposomes with a Solid-SupportedOctadecanol BilayerbyAmanda MusgroveB.Sc., University of Alberta, 2003M.Sc., The University of British Columbia, 2006A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinThe Faculty of Graduate and Postdoctoral Studies(Chemistry)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)October 2013? Amanda Musgrove 2013AbstractTransmembrane proteins and ion channels are a major target for new drug devel-opment. Incorporating them into sensors requires a method to produce stable, eas-ily modifiable solid-supported phospholipid bilayers. This thesis demonstrates amethod for using potential control on the electrode to mediate liposome adsorption,allowing them to interact with a previously deposited octadecanol layer throughpotential-created defects.Compression isotherms and electrochemical measurements were used to estab-lish the effect of the incorporation of a small amount of fluorescent dye on theoctadecanol layers. Using these fluorescently-labelled octadecanol layers, electro-chemical measurements both independently and coupled with in-situ fluorescencemeasurements were used to characterize the interaction of liposomes with theselayers under potential control. It was found that application of moderate potentials- more negative than the onset of defect formation but less than that required fordesorption of the layer - facilitated the effective incorporation of liposome materialinto the octadecanol bilayer. The length of time spent at the poration potential hadlittle effect on the degree of liposome interaction with the adsorbed layer. The in-corporation was seen as a change in the double-layer capacitance and the creationof small fluorescent structures in the layer after exposure to liposomes at the pora-tion potential. A shift in the characteristic desorption potential was also seen withliposome incorporation.Atomic force microscopy coupled in-situ with electrochemical control was alsoused to investigate the interaction of liposomes with the adsorbed octadecanol layer.The structure of the adsorbed layer was observed and with liposomes present insolution, the creation of three-dimensional structures similar in nature to those seenby fluorescence was noted. The incorporation of liposomes into the octadecanolwas shown to be easily controlled by application of an electrical potential, openinga path for a new method of producing supported lipid bilayers in-situ for biosensingapplications.iiPrefaceAll of the work presented henceforth was conducted in the Bizzotto group lab inthe Advanced Materials and Process Engineering Laboratory (AMPEL) at the Uni-versity of British Columbia, Point Grey campus.A previously published journal article (Potential dependent interaction of DOPCliposomes with an octadecanol covered Au(111) surface investigated using electro-chemical methods coupled with in-situ fluorescence microscopy, Langmuir) [1] hasbeen published that encompasses the material presented here in Chapter 5 and por-tions of the material in Chapter 4. This article may be accessed free of chargeat I was thelead investigator for this work, responsible for the major areas of concept forma-tion, experimental design, data collection and analysis, and manuscript composi-tion, with the exception of the work performed by the following authors. Colin R.Bridges was an undergraduate researcher in the laboratories of Drs. Bizzotto andSammis, and performed the synthesis of the BODIPY-C19-OH fluorophore used inthis investigation and composed the portion of the manuscript submitted as Sup-plementary Information (not included in this thesis). Glenn M. Sammis was the re-search supervisor of Colin R. Bridges, and contributed edits to the SupplementaryInformation portion of the manuscript. Dan Bizzotto was the supervisory authoron this project and was involved throughout the project in concept formation, dataanalysis, and manuscript composition.Colin R. Bridges also performed the data collection of the Langmuir isothermspresented in Section 3.1 under the supervision of myself and Dan Bizzotto. I wasresponsible for the remainder of the work presented in this thesis, including datacollection and analysis, experiment design, and manuscript composition. Dan Biz-zotto was the supervisory author on this project and was involved throughout theproject in concept formation, data analysis, and manuscript composition.iiiTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xNomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviiiAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xixDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xx1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Biosensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Phospholipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 Thin Films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.3.1 Surface Pressure and Langmuir Trough Measurements . . 41.3.2 Isotherms and Phase Diagrams for Thin Films . . . . . . 51.3.3 Multi-component Monolayers . . . . . . . . . . . . . . . 81.4 Solid-Supported Bilayers and Biosensors . . . . . . . . . . . . . 81.4.1 Formation of Solid-Supported Layers . . . . . . . . . . . 111.5 Electrochemical Background . . . . . . . . . . . . . . . . . . . 181.5.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 181.5.2 Octadecanol Electrochemistry . . . . . . . . . . . . . . . 251.5.3 Electrochemistry of DOPC and Vesicles . . . . . . . . . 331.6 Fluorescence Background . . . . . . . . . . . . . . . . . . . . . 42ivTable of Contents1.6.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 421.6.2 In-situ Fluorescence Imaging of Adsorbed layers at Elec-trode Surfaces . . . . . . . . . . . . . . . . . . . . . . . 521.7 Atomic Force Microscopy Background . . . . . . . . . . . . . . 561.7.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 561.7.2 Atomic Force Microscopy of Vesicles and Adsorbed Lay-ers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 611.7.3 In-situ Atomic Force Microscopy with Electrochemistry . 641.8 Scope of the Project . . . . . . . . . . . . . . . . . . . . . . . . 672 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . 712.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712.1.1 Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . 712.2 Liposome Formation . . . . . . . . . . . . . . . . . . . . . . . . 722.2.1 Chemical Stability of Liposomes . . . . . . . . . . . . . 722.3 Surface Pressure Measurements . . . . . . . . . . . . . . . . . . 732.3.1 Compression Isotherms and ESP . . . . . . . . . . . . . 732.3.2 In-situ Fluorescence Measurements . . . . . . . . . . . . 742.4 Preparation of Modified Electrodes . . . . . . . . . . . . . . . . 752.4.1 Liposome-coated Electrodes . . . . . . . . . . . . . . . . 752.4.2 Octadecanol-coated Electrodes: Single-touch (monolayer) 772.4.3 Octadecanol-coated Electrodes: Double-touch (bilayer) . 782.4.4 Multiple Depositions From Octadecanol Monolayers . . . 792.5 Electrochemical Methods . . . . . . . . . . . . . . . . . . . . . 792.5.1 Cyclic Voltammetry . . . . . . . . . . . . . . . . . . . . 802.5.2 Differential Capacitance . . . . . . . . . . . . . . . . . . 812.5.3 Incorporation of Liposomes . . . . . . . . . . . . . . . . 812.5.4 In-situ Fluorescence Methods . . . . . . . . . . . . . . . 822.6 Fluorescence Image Analysis . . . . . . . . . . . . . . . . . . . 832.7 Atomic Force Microscopy Methods . . . . . . . . . . . . . . . . 832.7.1 Instrumentation . . . . . . . . . . . . . . . . . . . . . . 832.7.2 Substrates . . . . . . . . . . . . . . . . . . . . . . . . . 842.7.3 Surface Modification . . . . . . . . . . . . . . . . . . . . 862.7.4 Imaging Conditions . . . . . . . . . . . . . . . . . . . . 86vElectrochemical Studies of Liposome Interaction...3 Effects of Fluorescent Dye on Octadecanol Monolayers . . . . . . . 883.1 Compression Isotherms . . . . . . . . . . . . . . . . . . . . . . 893.1.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 893.1.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 893.2 Electrochemical Characterization . . . . . . . . . . . . . . . . . 903.2.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 913.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 923.3 Fluorescence - Floating Layers and Compressions . . . . . . . . 933.3.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 963.3.2 Uncompressed Layers . . . . . . . . . . . . . . . . . . . 973.3.3 Compressions of Layers . . . . . . . . . . . . . . . . . . 993.4 In-situ Fluorescence with Electrochemistry . . . . . . . . . . . . 1033.4.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 1033.4.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1043.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1054 Electrochemical Studies of Liposome Interaction with Solid-SupportedOctadecanol Bilayers . . . . . . . . . . . . . . . . . . . . . . . . . . 1084.1 Electrochemistry of DOPC on Au(111) . . . . . . . . . . . . . . 1104.2 Interaction of DOPC with Floating Octadecanol Monolayers . . . 1124.3 Interaction of DOPC Liposomes with Octadecanol on Au(111) at0 V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1144.4 Effect of Poration Potential on Liposome - Octadecanol Interaction 1164.4.1 Experimental Design . . . . . . . . . . . . . . . . . . . . 1164.4.2 Capacitance During Potential Steps . . . . . . . . . . . . 1174.4.3 Desorption of the Modified Bilayers . . . . . . . . . . . . 1214.5 Effect of Poration Time on Liposome - Octadecanol Interaction . 1234.5.1 Experimental Design . . . . . . . . . . . . . . . . . . . . 1244.5.2 Capacitance During Application of Poration Potentials . . 1244.5.3 Desorption of the Modified Layers . . . . . . . . . . . . 1254.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1265 In-situ Fluorescence Studies of Liposome Interaction with Solid-SupportedOctadecanol Bilayers . . . . . . . . . . . . . . . . . . . . . . . . . . 1275.1 Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . 1295.2 Fluorescence at 0 V . . . . . . . . . . . . . . . . . . . . . . . . 131viIn-situ Fluorescence Studies of Liposome Interaction...5.3 In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 min . 1335.3.1 Fluorescence During Potential Steps . . . . . . . . . . . 1335.3.2 Fluorescence During Desorption of the Modified Layers . 1355.4 In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 min 1435.4.1 Fluorescence During Potential Steps . . . . . . . . . . . 1435.4.2 Fluorescence During Desorption of the Modified Layers . 1475.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1506 Atomic Force Microscopy Investigations of Octadecanol-coated Sur-faces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1526.1 Ex-situ Imaging of Octadecanol Monolayers on Au . . . . . . . . 1526.1.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 1536.1.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1536.2 Correlation with Fluorescence Imaging . . . . . . . . . . . . . . 1536.2.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 1556.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1566.3 In-situ Imaging under Potential Control . . . . . . . . . . . . . . 1576.3.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 1586.3.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 1596.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1687 Summary and Future Work . . . . . . . . . . . . . . . . . . . . . . 1707.1 Directions for Future Study . . . . . . . . . . . . . . . . . . . . 173References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176AppendicesA Liposome Characterization and Stability . . . . . . . . . . . . . . . 192A.1 Characterization by Thin Layer Chromatography . . . . . . . . . 192A.1.1 Experimental Methods . . . . . . . . . . . . . . . . . . . 192A.1.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . 193A.2 Characterization by Elastic Light Scattering . . . . . . . . . . . . 193B Fluorescence Calibration Factors . . . . . . . . . . . . . . . . . . . 196viiIn-situ Fluorescence Studies of Liposome Interaction...C Rolling-Ball Image Processing . . . . . . . . . . . . . . . . . . . . . 197viiiList of Tables3.1 Average red:green (R/G) ratios for octadecanol monolayers con-taining 1 mol% and 3 mol% BODIPY-C19-OH. . . . . . . . . . . . 1024.1 Equilibrium surface pressure (ESP) of octadecanol and mixed oc-tadecanol/DOPC monolayers. . . . . . . . . . . . . . . . . . . . . 112A.1 Mean particle sizes for liposome solutions as measured using dy-namic light scattering. . . . . . . . . . . . . . . . . . . . . . . . 195B.1 Conversion factors to kcts/sec used for fluorescence images. . . . 196ixList of Figures1.1 dioleoyl phosphatidylcholine (DOPC) . . . . . . . . . . . . . . . 31.2 Differences in aggregation shapes of a) single-tailed lipids and b)double-tailed lipids (e.g. phospholipids). . . . . . . . . . . . . . . 31.3 Schematic diagram of a Langmuir trough. . . . . . . . . . . . . . 51.4 Langmuir isotherm for octadecanol at 20?C . . . . . . . . . . . . 71.5 Phase diagram for octadecanol . . . . . . . . . . . . . . . . . . . 71.6 Brewster angle microscopy images of floating octadecanol mono-layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.7 Schematic of miscibility of mixed monolayers . . . . . . . . . . . 101.8 Collapse pressures of immiscible and miscible layers. . . . . . . . . 101.9 Deposition techniques for forming solid-supported bilayers fromfloating monolayers. . . . . . . . . . . . . . . . . . . . . . . . . . 121.10 Bilayer arrangements formed from Langmuir-Blodgett depositions. . . 121.11 Proposed mechanisms of vesicle bursting and fusion to form sup-ported lipid bilayers. . . . . . . . . . . . . . . . . . . . . . . . . . 151.12 Depiction of PEG-PE containing vesicle and bilayer. . . . . . . . . 171.13 Examples of different methods of forming bilayers on templatedsurfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191.14 Schematic model of the electrode-solution interface. . . . . . . . . 201.15 RC Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211.16 Effect of the increase of bulk activity of the surfactant (a: lowest,d: highest) on the interfacial tension. . . . . . . . . . . . . . . . . . 231.17 Differential capacitance-potential curve of a HMDE in: (1) a 0.5 MNa2SO4, (2) 0.5 M Na2SO4 solution saturated with camphor.. . . . 241.18 Differential capacity of a Hg electrode in 1 M KNO3. . . . . . . . . 261.19 Capacitance behaviour of octadecanol on a HMDE. . . . . . . . . . 27xList of Figures1.20 Cyclic voltammogram (a) and differential capacitance (b - dashedline: negative scan. solid line: positive scan) of an octadecanolmonolayer on Au(111) in 0.05 M KClO4. . . . . . . . . . . . . . . 291.21 Effect of varying initial surface pressure on differential capacitancescans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291.22 Elastically scattered light on octadecanol monolayer Au(111). . . . 301.23 Changes in octadecanol chain tilt angle as determined by PM-IRRAS. . . 311.24 Proposed mechanism of desorption and readsorption of a bilayer ofoctadecanol on Au(111). . . . . . . . . . . . . . . . . . . . . . . . 321.25 Differential capacitance of Langmuir-Schaefer, X-type, and Y-typebilayers of octadecanol on Au(111). . . . . . . . . . . . . . . . . . 331.26 Differential capacitance of L-?-lecithin (open circles) and DL-?-lecithin (closed circles) monolayers . . . . . . . . . . . . . . . . . . 341.27 Differential capacitance of a DOPC monolayer adsorbed from thegas|solution interface on a mercury drop electrode. . . . . . . . . 351.28 Differential capacitance desorption scans of DOPC (dotted line)and DMPC (solid line) bilayers formed from vesicles in solution. . 361.29 Cartoon representation of the behaviour of DMPC on an Au elec-trode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371.30 Adhesion signal for multilamellar DOPC vesicles adhering to amercury electrode. . . . . . . . . . . . . . . . . . . . . . . . . . 391.31 Differential capacitance of DOPC on a mercury electrode. . . . . . 401.32 Proposed mechanism for liposome interaction with an existing mono-layer of DOPC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411.33 Partial energy diagram for a photoluminescent system. . . . . . . 431.34 Application of FRET to RNA sensing . . . . . . . . . . . . . . . . 451.35 Proposed structures of the BODIPY dimers. . . . . . . . . . . . . . 451.36 Excited state lifetime of Eu3+in air near a silver mirror surface. . . 461.37 Oscillating field of dipole interacting by interference with a planemirror. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461.38 Fluorescence sensor for DNA binding. . . . . . . . . . . . . . . . . 481.39 Components of fluorescence microscopy. . . . . . . . . . . . . . 491.40 Light cone of a microscope objective. . . . . . . . . . . . . . . . 501.41 Rayleigh criterion for spatial resolution. . . . . . . . . . . . . . . . 51xiList of Figures1.42 Fluorescence imaging of an octadecanol monolayer with 3 mol %DilC18(5) dye on a Au(111) electrode. . . . . . . . . . . . . . . . 541.43 Fluorescence images of an octadecanol monolayer before adsorp-tion (a) and the same region after adsorption . . . . . . . . . . . . . 551.44 Tip-sample interactions at specific points in a force-curve cycle forAFM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 571.45 Frequency and phase response of a cantilever modeled as a dampedharmonic oscillator. . . . . . . . . . . . . . . . . . . . . . . . . . . 591.46 AFM images of a polycrystalline gold sample. . . . . . . . . . . . 601.47 AFM height and phase images of a polyvinyl alcohol thin film. . . . 601.48 AFM image (taken in contact mode) of a mica surface modifiedwith a partial-double-bilayer of phosphatidylethanolamine, with thecentre of the imaging area additionally modified by a high-forceimaging scan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621.49 Force vs distance curve of an AFM tip placed directly onto an eggphosphatidylcholine liposome. . . . . . . . . . . . . . . . . . . . . 631.50 Cartoon schematic of a lipid raft . . . . . . . . . . . . . . . . . . . 641.51 AFM images of mixed sphingomyelin/DOPC monolayers (1:1 mol/mol)containing various amounts of cholesterol. . . . . . . . . . . . . . 651.52 Atomic force microscopy images of N-dodecyl-N,N-dimethyl-3-ammonio-1-propanesulfonate. . . . . . . . . . . . . . . . . . . . . 651.53 Atomic force microscopy image (MAC mode, in 0.1 M NaF) of adimyristyl phosphatidylcholine layer under potential control. . . . . 661.54 Illustration of the proposed hybrid bilayer formation process. . . . . 681.55 Differential capacitance scans during desorption of a DOPC bilayer(top frame) and octadecanol bilayer (bottom). . . . . . . . . . . . . 692.1 Epi-fluorescence microscope configuations used for in-situ fluores-cence imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . 752.2 Transmission characteristics of filter cubes used for fluorescencemeasurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . 762.3 Schematics of electrochemical cells used, shown with Au(111) elec-trode in hanging meniscus configuration. . . . . . . . . . . . . . . 772.4 Potential step profiles used in testing liposome incorporation. . . . 822.5 Assembly of AFM electrodes used for imaging. . . . . . . . . . . . 85xiiList of Figures3.1 Compression isotherms for monolayers of octadecanol containing0, 1, 3, 5, and 8 mol% BODIPY-C19-OH. Film pressure range plot-ted is limited to the region below the minimum collapse pressurefor all layers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 903.2 Equilibrium spreading pressure (ESP) and minimum capacitancevalues for octadecanol monolayers containing 0, 0.5, 1, 3, 5, and 8mol % BODIPY-C19-OH fluorophore. . . . . . . . . . . . . . . 913.3 Cyclic voltammograms (top) and differential capacitance scans (bot-tom) of octadecanol monolayers containing various concentrationsof BODIPY-C19-OH on Au(111). . . . . . . . . . . . . . . . . . 943.4 Cyclic voltammograms (top) and differential capacitance scans (bot-tom) of octadecanol bilayers containing various concentrations ofBODIPY-C19-OH on Au(111). . . . . . . . . . . . . . . . . . . . 953.5 Fluorescence images of a floating octadecanol monolayer contain-ing 1 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 973.6 Fluorescence images of a floating octadecanol monolayer contain-ing 3 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. . . 983.7 Fluorescence images of a floating octadecanol monolayer contain-ing 1 mol% BODIPY-C19-OH fluorophore in a Langmuir trough.Images taken with barriers partially closed. . . . . . . . . . . . . . 1003.8 Fluorescence images of a floating octadecanol monolayer contain-ing 3 mol% BODIPY-C19-OH fluorophore in a Langmuir trough.Images taken with barriers partially closed. . . . . . . . . . . . . . 1013.9 In-situ fluorescence and differential capacitance measurements of abilayer of octadecanol containing BODIPY-C19-OH fluorophore. . . 1064.1 Differential capacitance scans during desorption of a DOPC bilayer(top frame) and octadecanol bilayer (bottom). . . . . . . . . . . . . 1094.2 Differential capacitance measurements of DOPC on Au(111) . . . . 1104.3 Surface pressure (?) during exposure to liposomes. . . . . . . . . 1134.4 Differential capacitance of octadecanol bilayers at 0 V/SCE duringexposure to DOPC liposomes for 60 min. . . . . . . . . . . . . . . 1154.5 Potential step profiles used in testing liposome incorporation. . . . 117xiiiList of Figures4.6 Differential capacitance of octadecanol bilayers on Au(111) withand without liposomes in solution during application of potentialsteps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1184.7 Replicate measurements of octadecanol capacitance with and with-out liposomes in solution during application of the -0.6 V and -0.4 V/SCE poration potential. . . . . . . . . . . . . . . . . . . . . 1194.8 Capacitance during a potential sweep to desorption (+0.15 to -0.8 V/SCE) of octadecanol layers subjected to the potential profilesin Figure 4.6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1224.9 Differential capacitance of octadecanol bilayers on Au(111) withand without liposomes in solution during application of -0.4 V/SCEfor various times. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1244.10 Capacitance during a potential sweep to desorption (+0.15 to -0.8 V/SCE) of octadecanol layers subjected to the potential profilesin Figure 4.9. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1255.1 Cartoon illustration of a possible mechanism of increasing fluores-cence during liposome incorporation. . . . . . . . . . . . . . . . . 1285.2 Average fluorescence intensity and differential capacitance mea-surements for layers held at 0 V/SCE. . . . . . . . . . . . . . . . . 1305.3 Spectroelectrochemical behaviour of adsorbed octadecanol bilay-ers with and without liposomes in solution during application of apotential profile with a 1 min poration step. . . . . . . . . . . . . . 1365.4 Representative fluorescence images taken during application of apotential profile with a 1 min poration step. . . . . . . . . . . . . . 1375.5 Spectroelectrochemical behaviour of adsorbed octadecanol bilay-ers with and without liposomes in solution during desorption of thelayer after modification by a 1 min poration potential profile. . . . . 1405.6 Representative fluorescence images taken during application of apotential profile with a 1 min poration step. . . . . . . . . . . . . . 1415.7 Fluorescence images of Figure 5.6 after background subtraction bya 50 pixel radius rolling ball filter. . . . . . . . . . . . . . . . . . . 1425.8 Spectroelectrochemical behaviour of adsorbed octadecanol bilay-ers with and without liposomes in solution during application of apotential profile with a 15 min poration step. . . . . . . . . . . . . 144xivList of Figures5.9 Representative fluorescence images taken during application of apotential profile with a 15 min poration step. . . . . . . . . . . . . 1455.10 Spectroelectrochemical behaviour of adsorbed octadecanol bilay-ers with and without liposomes in solution during desorption of thelayer after modification by a 15 min poration potential profile. . . . 1485.11 Representative fluorescence images taken during application of apotential profile with a 15 min poration step. . . . . . . . . . . . . 1495.12 Fluorescence images of Figure 5.6 after background subtraction bya 50 pixel radius rolling ball filter. . . . . . . . . . . . . . . . . . . 1496.1 AFM image of a n octadecanol monolayer containing 3 mol% BODIPY-C19-OH on the Au(111) facet of a bead electrode. Images wereacquired in ACAFM mode in air. . . . . . . . . . . . . . . . . . 1546.2 AFM image of an octadecanol monolayer containing 3 mol% BODIPY-C19-OH on a Au/mica substrate. Image acquired in MAC mode inair. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1546.3 AFM topography image (left) of the Au(111) facet with octade-canol monolayer. (Right) Mask of major topographical featuresfrom the AFM image. . . . . . . . . . . . . . . . . . . . . . . . . 1566.4 Fluorescence image of the region surrounding the area imaged inFigure 6.3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1576.5 In-situ AFM images of an octadecanol bilayer containing 3 mol%BODIPY-C19-OH: topography channel. Potentials are measuredvs. an Au bead reference. . . . . . . . . . . . . . . . . . . . . . . 1606.6 In-situ AFM images of an octadecanol bilayer containing 3 mol%BODIPY-C19-OH: phase channel. Potentials are measured vs. anAu bead reference. . . . . . . . . . . . . . . . . . . . . . . . . . 1616.7 Example features of each category from the AFM topography im-ages of octadecanol bilayers. Images are cropped from those inFigure 6.5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1626.8 In-situ AFM images of an octadecanol bilayer containing 3 mol%BODIPY-C19-OH and liposomes in solution: topography channel.Potentials are measured vs. an Au bead reference. . . . . . . . . . 163xvList of Figures6.9 In-situ AFM images of an octadecanol bilayer containing 3 mol%BODIPY-C19-OH and liposomes in solution: phase channel. Po-tentials are measured vs. an Au bead reference. . . . . . . . . . . 1646.10 Example features that were present in the initially deposited layerof each category from the AFM topography images. . . . . . . . . 1656.11 Example features that appeared during liposome adsorption fromthe AFM topography images. . . . . . . . . . . . . . . . . . . . . 1666.12 Fluorescence images taken during the application of potential stepsafter background subtraction by application of a 50 pixel rollingball filter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1687.1 Cartoon schematic of protein incorporation into a raised liposomalbilayer structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1717.2 A possible procedure for creating multifunctional sensors in-situ ina microfluidic cell. . . . . . . . . . . . . . . . . . . . . . . . . . . 174A.1 TLC plates taken at various points during the lifetime of a liposomesolution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194C.1 False-color fluorescence images processed with a rolling-ball filterof various ball sizes. False-color scale is the same for all images,and all images are 150 x 150 ?m. . . . . . . . . . . . . . . . . . 198xviNomenclature? dielectric constant?0 permittivity of free space? Surface excess (Gibbs excess)? Surface (or interfacial) tension? chemical potential? Surface pressure? Charge density12-AS 12-(9-anthroyloxy) stearic acidAC Alternating currentAC Intermittent contact AFM imaging modeAFM Atomic force microscope/microscopyBAM Brewster angle microscopyBODIPY-C19-OH 4,4-difluoro-1,3,5,7-tetramethyl-8-(18-octadecanol)-4-bora-3a,4a-diaza-s-indaceneCV Cyclic voltammetry/voltammogramDilC18(5) 1,1?-dioctadecyl-3,3,3?,3?-tetramethylindodicarbocyanine perchlorateDMPC Dioleoyl myristoyl phosphatidylcholineDNA Deoxyribonucleic acidDOPC dioleoyl phosphatidylcholinexviiNomenclatureEQCM Electrochemical quartz crystal microbalanceESP Equilibrium surface pressureFRET F orster Resonance Energy TransferHMDE Hanging mercury drop electrodeHMDE Hanging mercury drop electrodeIHP Inner Helmholtz Planekcts/sec kilocounts per secondMAC Magnetic AC (AFM vibrational imaging mode)NA Numerical apertureOHP Outer Helmholtz PlanePM-IRRAS Polarization modulation infrared reflection absorption spectroscopypzc Potential of zero chargeQCM Quartz crystal microbalanceRMS Root mean squareSAM Self-assembled monolayerSCE Saturated caolmel electrodeSTM Scanning tunneling microscopyTLC Thin layer chromatographyxviiiAcknowledgementsMany people have helped me along the way - but some special thank yous arewarranted.First, to my supervisor, Dr. Dan Bizzotto. Your guidance and advising has beeninvaluable, and a makes a lab worth coming back to.To the members of the Bizzotto lab group, past and present - Dr. Jeff Shepherd,Dr. Robin Stoodley, Dr. Aya Sode, Jeffrey Murphy, Jannu Casanova Moreno,Landis Yu, Santa Maria Gorbunova, and Isaac Martens. You are a great group towork with, and your discussions, friendship, and teatimes have been immenselyhelpful, both in the lab and out.All of the glass electrochemical cells used in this work were created by thedepartment?s glassblower, Brian Ditchburn. Thanks Brian - this work, especiallywith the fluorescence cells, literally could not have happened without your skill.Thank you to Dr. Guillaume Bussiere for the use of your Langmuir trough toacquire the data shown in Chapter 3, and for assisting Colin Bridges in his mea-surements on the same.A special thank you to Lyndsey Earl - your friendship has brightened my timehere, and your scientific discussions (notably on optimizing the TLC analysis) havegotten me unstuck many a time.Thank you to my family - my husband, parents, and sisters - your love andsupport brought me here and kept me going.Financial support from the Agnes and Gilbert Hooley Scholarship in Chemistryis gratefully acknowledged.xixDedicationThis thesis is dedicated to my husband Laurent - Thank you. You have supportedme through more schooling than is really right, and your love, support, and under-standing have made me a better person through it all.xxChapter 1IntroductionThe use of solid-supported lipid bilayers as a platform for developing biosensors isa growing field of study, especially as up to 60% of new drug targets are membraneproteins [2]. The ability to support these proteins in their natural environment isessential to create devices that rely on monitoring their functionality. This thesiswill describe a novel process for producing a solid-supported bilayer assembly fordevelopment as a biosensor platform. The rationale and perceived challenges forthis project [3] will be outlined below, along with a brief introduction to the neces-sary theory and a review of the state of the art regarding the techniques used in thisstudy.1.1 BiosensorsFrom the now-ubiquitous electrochemical glucose sensor [4] to detection of newdrug targets, sensors for biological molecules are an actively developing area. Inits broadest definition, a biosensor is a device that is able to detect chemical com-pounds based on biochemical reactions and transduces the reaction or binding eventinto a signal that is readily measured. [5] The reaction used for detection in the sen-sor could involve the analyte interacting with a specific ligand, such as an antibodyor DNA aptamer designed to interact with the target compound. Enzymes andpeptides as detection targets present a special challenge as they often have a com-plex three dimensional structure that requires a carefully controlled environmentto maintain - including the availability of a phospholipid bilayer similar to a cellmembrane for some proteins. Happily, with careful design these sensors can beincredibly effective, even detecting concentrations in the attomole range [6], andeven down to the single molecule level. [7]Biosensors based on lipid bilayers supported on a planar substrate are morerobust than those based on other bilayer models, such as a black lipid membrane.The stability of the supported layers may be further enhanced by forming tethered11.2. Phospholipidsbilayers wherein a molecule, such as polyethylene glycol or a thiolipid, acts as amediator between the bilayer and the substrate and both anchors the bilayer to thesurface and provides a cushioning space to ensure room for water and ion flowbetween the electrode and the bottom leaflet of the bilayer. [8] Maintaining a watergap between the supported bilayer and the electrode surface is also important forincorporating proteins with large extra-membranal domains, as they may denatureif they come into direct contact with the electrode surface. [9]Embedding a protein or ion channel in a supported lipid membrane providessensor functionality. Since ions cannot pass through the hydrophobic core of the bi-layer without a carrier molecule, biomolecules that transport ions across the mem-brane will generate an easily detectable perturbation in the electrochemical charac-teristics of the system. Incorporation of tetrachloro-o-benzoquinone or tetrachloro-p-benzoquinone in electrode-supported bilayers creates a pH sensitivity [10], whilesensitivity to ions can be created by use of ion carriers such as valinomycin or crownethers. [11] Ions may also pass through a lipid bilayer via an ion channel. Theseare either voltage-gated or ligand-gated, and can be based on natural [12?15] orsynthetic [16] ion channels. Gramicidin was among the first ion channels used toproduce an electrochemical sensor [17] by measuring the change in conductanceacross the membrane. This design has been developed to produce a sensor for in-fluenza virus designed for point-of-care analysis. [18] Other peptides have beenincorporated into solid supported membranes [19?22] as well as ion channels [23?25]. Although solid supported bilayers as a platform are less developed than othermodels, they are the only current option for sensing with membrane proteins that issufficiently stable and suited for making robust, commercially viable sensors. [26]1.2 PhospholipidsThe membrane that surrounds biological cells is composed primarily of a mixed-phospholipid bilayer, embedded with proteins and carbohydrates, especially choles-terol. Liposomes resemble the cell membrane, though proteins and carbohydratesare not usually added and the phospholipid composition is controlled, usually con-sisting of only one or two types of molecule.There are two main classifications of phospholipids: glycerophospholipids andsphingomyelins. The focus of this study is on liposomes formed from glycerophos-pholipids. Sphingomyelins, though an important membrane component, will not be21.2. PhospholipidsFigure 1.1: dioleoyl phosphatidylcholine (DOPC)Figure 1.2: Differences in aggregation shapes of a) single-tailed lipids and b)double-tailed lipids (e.g. phospholipids).discussed further. Glycerophospholipids consist of a glycerol backbone to whichare attached two esterfied fatty acids at the C1 and C2 positions, and a phosphategroup on C3. Another group may be linked to the phosphate group as well. Onecommon phospholipid, dioleoyl phosphatidylcholine (DOPC), is depicted in Figure1.1. Note that this particular phospholipid, though uncharged overall, is zwitteri-onic at neutral pH, enhancing the hydrophilicity of the phosphate head group, whilethe hydrocarbon ?tails? are hydrophobic. This amphiphilic behavior causes phos-pholipids to aggregate when placed in aqueous solution. However, unlike someother surfactants, which tend to form spherical micelles, the more cylindrical crosssection of a phospholipid molecule favors a planar structure - the bilayer - as shownin Figure 1.2. Liposomes may be unilamellar, having only one bilayer, or multi-lamellar, being composed of several nested spherical bilayers. They may also beclassified by size, ?small? vesicles being on the order of 20 - 200 nm, ?large? vesi-cles 200 - 1000 nm, and ?giant? liposomes larger than 1 ?m, often on the samescale as a typical cell. [27]For most phospholipids, the bilayer is a liquid crystalline structure at biologi-cally relevant temperatures [28], though it does have at least one phase transition,depending on the type of phospholipid involved. All such bilayers undergo a ?melt-31.3. Thin Filmsing? transition from a gel-like state to a more liquid state at a characteristic temper-ature Tm. This transition is highly cooperative in nature, caused by a disorderingof the hydrocarbon tails and a corresponding increase in area per molecule. Thetemperature of the phase change is directly affected by the length of the tails, aswell as the degree of unsaturation. Strain on the bilayer, such as bending into aspherical shape in a liposome or electrostriction, can also affect the phase transi-tion, lowering the Tm. [28] Bilayers tend to have many structural defects, as thehydrocarbon tails favour a different lattice than the polar head groups. The two hy-drocarbon tails may also favor different orientations, further frustrating the layer.These defects may be stabilized by attraction of solutes into the bilayer structure,or orientation of molecules out of the bilayer plane, relaxing the defect. [29]1.3 Thin FilmsThe behaviour of thin films at the gas|solution interface is important for the depo-sition methods used in this work, as well as for characterizing the structure andbehaviour of insoluble surfactants. Thin films, especially monomolecular films andbilayers, are used extensively in forming the biosensors discussed above, and anunderstanding of their behaviour is essential for designing and creating such tools.1.3.1 Surface Pressure and Langmuir Trough MeasurementsThe method of using a trough with movable barriers to examine the effects of thinfilms on surface pressure were developed first by Pockels [30] and further refined byothers, notably Langmuir and Blodgett who developed a technique for depositingmonolayers and multilayers from the gas|solution interface onto a solid substrate.[31?33] Typically, the behaviour of a thin film of insoluble surfactant (such asoctadecanol, a fatty alcohol) is measured by the use of a Langmuir Trough (Figure1.3). Briefly, the trough consists of a reservoir filled with solvent and bound byone or two movable barriers. A microbalance with Wilhelmy plate is arrangedso that the plate is in contact with the solvent between the barriers, and a smallamount of the surfactant molecule is introduced to the gas|solution interface. Oncethe surfactant film has formed, the film pressure ? can be determined from the netforce on the Wilhelmy plate. The net downward force for a plate of density ?p,known width w, length l and thickness t, immersed in a solvent of density ?s to a41.3. Thin FilmsFigure 1.3: Schematic diagram of a Langmuir trough.depth h, is the balance of gravity, surface tension, and buoyancy:F = ?pglwt +2?(t +w)cos? ??sgtwh (1.1)where ? is the contact angle of the solvent with the plate, g the acceleration due togravity, and ? the surface tension of the liquid. By taking the difference betweenthe clean subphase and the surfactant-modified surface, the film pressure of thesurfactant can be found:? = ?water ? ? f ilm =??F2(t +w)???F2w(1.2)where the final form of the equation holds when the plate thickness is negligiblecompared to the width, as is generally true for plates made from paper or similarthin materials. By monitoring the film pressure as the barriers are brought closertogether, reducing the area available for the surfactant, insight is gained into thestructure and interaction of the surfactant molecules as their surface concentrationincreases.1.3.2 Isotherms and Phase Diagrams for Thin FilmsThe plot of ? vs. molecular area for compression of a surfactant at a constanttemperature is called a Langmuir isotherm. A typical Langmuir isotherm for oc-tadecanol is shown in Figure 1.4. In this isotherm, several regimes of responseto changes in area can be clearly seen, based on changes in linearity and slope ofthe isotherm. These follow the generally predicted trends for compression of along-chain alcohol. Described using the phase descriptions introduced in [34] and51.3. Thin Filmssummarized in [33, 35], up to six distinct phases are distinguishable in long-chainalcohols. According to the phase diagrams predicted for long-chain molecules col-lected in [36] and further investigated in [35] and [37], octadecanol does not haveaccess to this lowest density phase at the film pressures and temperatures mea-sured. Instead the lowest density phases observed in the long-chain alcohols is theL2 liquid-condensed phase. This phase is liquid-crystal like with some translationalorder, although the layer remains compressible. At room temperature, higher filmpressure will result in the LS phase, the so-called ?superliquid? phase. This phaseis also a liquid-crystal like mesophase, but with higher order - a hexagonal unitcell, and the aliphatic chains oriented perpendicularly to the gas|solution interface.At lower temperatures, the solid (S) and compact solid (CS) states are accessible.Despite the term ?solid? used in these phases, x-ray diffraction studies [38] suggestthat these phases are also more liquid-crystal like than true solid phases. The em-pirical phase diagram for a monolayer of octadecanol is shown in Figure 1.5. In theisotherm of Figure 1.4, the transitions between these phases are visible as changesin the slope, or ?kinks? in the isotherm plot. Eventually, as compression increases,the layer begins to buckle and form multilayers, collapsing, and the film pressureremains at a maximum value as the area decreases. Typically, this collapse pressureis at or above the equilibrium spreading pressure (ESP) of the monolayer - the filmpressure it will naturally obtain when a film is formed with an excess of surfactantin a fixed area. For an uncollapsed monolayer at its ESP, any excess surfactant ma-terial will be located in small crystallites or lenses of multilayer interspersed withthe monolayer. Octadecanol has a well-characterized ESP typically at 32-35 mN/m[33, 35], which suggests that it will be in the LS phase - a high-density liquid crystal- when formed at equilibrium conditions at room temperature.As the terminology used for the phases suggests, the monolayers of octadecanoland similar surfactants formed at the gas|solution interface are not perfectly crys-talline or defect free. As a demonstration of the defects present in a compressedoctadecanol layer, Slevin et al. [39] used an ultramicroelectrode placed facing up-wards beneath an octadecanol layer in a Langmuir trough. The interfacial resistanceof the oxygen transfer across the air|solution interface was measured. From a pres-sure of 5 mN/m to approximately 50 mN/m (the point of collapse), oxygen transferacross the monolayer was detected. The rate of the transfer depended linearly withthe molecular area, and interfacial resistance increased as the surface pressure in-creased, fitting a model of the monolayer containing defects that permit oxygen61.3. Thin FilmsFigure 1.4: Langmuir isotherm for octadecanol at 20?C. Phase symbols are as de-fined in Figure 1.5. Reprinted from [35] with permission from Elsevier.Figure 1.5: Phase diagram for octadecanol. Phase abbreviations: S - solid, CS -compact solid, LS - superliquid, L2- liquid condensed. Reprinted from [35] withpermission from Elsevier.71.4. Solid-Supported Bilayers and Biosensorstransfer into the subphase that are gradually removed as the layer is compressed.Brewster angle microscopy (BAM) has shown that even pure octadecanol mono-layers have regions of coexisting phases at a range of film pressures. [40] Higherfilm pressures (corresponding to lower molecular area) have larger regions of condensed-phase octadecanol, but even near the molecular area at collapse (~19 ??/molecule),regions of uncompressed layer exist, as seen in Figure 1.6e. The presence of de-fects in octadecanol monolayers formed in a Langmuir trough may also be inferredfrom capacitance measurements, as discussed in Section1. Multi-component MonolayersMixed monolayers may be formed that contain more than one type of surfactantmolecule. If the compounds in the mixture are completely miscible, the compo-sition across the floating monolayer will be even (Figure 1.7), and the collapsepressure of the film will vary depending on the film composition. For compoundsthat are completely or partially immiscible, pure regions of each component willexist (Figure 1.7), and the film will collapse at the lowest collapse pressure of theindividual components, as regions of this component collapse. [33] An example ofthis is shown in Figure 1.8 for mixtures of C16 and C24 fatty acids. The immisciblecombination of C16 and C24 shows a constant collapse pressure near the lower ofthe two compounds? ESP for all mixtures containing both compounds. A plateaualso appears in the isotherm at an intermediate pressure, believed to be due to the?squeezing out? of the lower-pressure component. [41] For miscible compounds,such as C16 and C20 fatty acids in Figure 1.8, the collapse pressure increases withthe increasing amount of C20 to a maximum at 100% C20. The similar segregationof long chain (16 to 22-carbon) alkyl alcohols has been observed by AFM, [42] con-firming that alcohol combinations where the chain lengths are less than 6 carbonsdifferent tend to be miscible, with no segregated domains visible by AFM. Greaterdifferences in chain length lead to domain formation visible by ?/area isothermmeasurements and AFM.1.4 Solid-Supported Bilayers and BiosensorsThin monomolecular films of the type described in Section 1.3 can be studied di-rectly, and phospholipid bilayers as ?black lipid membranes? can be formed across81.4. Solid-Supported Bilayers and BiosensorsFigure 1.6: Brewster angle microscopy images of floating octadecanol mono-layers at various film pressures. Light areas correspond to condensed phase re-gions, and dark regions are less-condensed ?gaseous? phase. a) 120 ??/moleculeb) 100 ??/molecule c) 60 ??/molecule d) 30 ??/molecule e) 20 ??/molecule.Reprinted with permission from [40]. Copyright 1996 American Chemical Society.91.4. Solid-Supported Bilayers and BiosensorsFigure 1.7: Schematic of miscibility of mixed monolayers. (a) Components com-pletely miscible (b) Components partially miscible (small domains). (c) Compo-nents completely immiscible (complete phase segregation).(a) Collapse and transition pressures for the mis-cible mixtures of C16 and C20 fatty acids.(b) Collapse, transition, and plateau pressures forthe immiscible mixtures of C16 and C24 fattyacids.Figure 1.8: Collapse pressures of immiscible and miscible layers. Reproducedfrom [43] with kind permission from Springer Science and Business Media.101.4. Solid-Supported Bilayers and Biosensorsa small hole and used to investigate transport processes, as summarized in [44].However these membranes are delicate, and have a short lifetime. By associatingthe bilayer onto a solid support, greater stability of the layer and more facile meth-ods of formation can be accessed, making this platform a popular topic of recentstudy.1.4.1 Formation of Solid-Supported LayersSolid-supported layers can be broadly categorized into two types: chemisorbed andphysisorbed layers. Chemisorbed species are more strongly attached to the solidsupport, generally involving formation of a chemical bond between the adsorbateand surface. A thiol self-assembled monolayer (SAM) on gold, attached by a gold-sulfur bond, is a typical example of a chemisorbed species. Physisorbed speciesare held at the solid support only by van der Walls attraction between the adsorbateand surface, and are therefore generally less robust than chemisorbed layers, asevidenced by the typical enthalpy of adsorption for physisorption being on the orderof 10 times less than for chemisorption. [45] Although solid-supported layers areeasily formed by deposition from a Langmuir trough or similar, it is also simple touse self-assembly from a solution of the adsorbate to form chemisorbed layers, andin some cases (such as self-assembly from vesicles) to self-assemble physisorbedlayers.Langmuir-Blodgett depositionSolid-supported mono-, bi-, and multi-layers may be conveniently formed from afloating surfactant monolayer of the type described in Section 1.3 by the processof Langmuir-Blodgett deposition. [31, 33] In this technique, a floating monolayeris formed and compressed to the desired film pressure in a Langmuir trough, whilethe solid support is suspended vertically above the air|solution interface. The solidsupport is then slowly dipped through the interface and the monolayer supportedthereon, depositing the monolayer onto the solid support. Figure 1.9a depicts thisprocess, as well as a variant where the solid support is initially in the subphasebelow the interface, and is pulled up to deposit the monolayer.The surface tension of the interface and direction of motion of the solid sup-port during the monolayer transfer onto the solid support will influence the finalorientation of the surfactant molecules; for a monolayer of surfactant on an aque-111.4. Solid-Supported Bilayers and Biosensors(a) Langmuir-Blodgett deposition using the ?pull?method (left) and ?push? method (right).(b) Bilayer formation byLangmuir-Schaefer deposi-tion. In this case, the firstleaflet of the bilayer haspreviously been deposited,and the outer leaflet is beingadded.Figure 1.9: Deposition techniques for forming solid-supported bilayers from float-ing monolayers. Adapted with permission from [46]. Copyright 2004 AmericanChemical Society.Figure 1.10: Bilayer arrangements formed from Langmuir-Blodgett depositions.Left: X type. Center: Y type. Right: Z type. Adapted with permission from [46].Copyright 2004 American Chemical Society.ous subphase, the hydrophobic tail groups will be facing the upper surface of themonolayer, so a solid support pushed downwards through the interface will forma layer with the tail groups closest to the support. This compression and dippingprocedure can be repeated multiple times to form multi-layers of surfactant on thesolid support. By varying the direction of the dipping, different combinations ofmolecular orientations can be achieved [41], as summarized in Figure 1.10. Themost common type of multi-layer deposition, where the leaflets are deposited withalternating ?push? and ?pull? directions, is the Y-type, and results in layers thathave leaflets facing head-to-head and tail-to-tail. Layers of X and Z type have ahead-to-tail leaflet orientation, and result from all-push and all-pull layer forma-tion, respectively.121.4. Solid-Supported Bilayers and BiosensorsA related method for depositing monolayers onto solid supports is the Langmuir-Schaefer deposition (Figure 1.9b). This method uses a floating monolayer as doesLangmuir-Blodgett deposition, however rather than moving the substrate throughthe interface, it is brought in parallel to the interface and gently touched to thesurface, then lifted away. The process may be repeated to form multi-layers, butbecause there is only one possible orientation for the surfactant molecules in eachdeposition, all multilayers formed in this way will necessarily be X-type.A special case of mixed layers are those of alkyl alcohols mixed with alkylth-iols. Although forming these mixed monolayers on the liquid|gas interface is lessfavourable because the thiol group is less polar than the alcohol group, when de-posited onto a suitable substrate (such as gold), the added attraction from thechemisorption of the thiol fraction causes the deposited layer to be more stable.With the addition of octadecanethiol to Langmuir monolayers of octadecanol, ithas been shown [47] that with increasing fraction of octadecanethiol, the mono-layer had a decreasing collapse pressure, but when deposited onto a gold electrode,higher octadecanethiol content resulted in a lower capacitance and lower signalfrom a redox probe, suggesting fewer defects in the layer. See Section 1.5.2 forfurther discussion on this topic.Self-assembly of chemisorbed monolayersA simple and accessible method for forming monolayers on solid substrates is self-assembly from a solution of the monomer. This approach has been taken with manyclasses of molecules and suitable substrates, such as silanes on glass [48] and onoxidized metal surfaces such as aluminum and titanium, [49, 50] diazonium on sili-con, [51] and perhaps the most studied system, thiols and related sulfur compoundson metals, especially gold. Thiol SAMs are typically formed by immersing thesubstrate in a solution of the thiol compound for a given time (from a few hoursto several days [52]) and allowing the layer to self-assemble onto the surface. Theprocess can be aided by application of an electrical potential [53], encouraging theoxidation of the thiol into the metal-thiolate complex.Thiol SAMs are generally robust, and the method has been used for attach-ing important biological molecules, both by incorporating a thiol group onto themolecule, as for DNA-cyclodextrin, peptides, and other compounds [52], or mak-ing use of naturally accessible sulphur in cysteine for attaching suitable proteins toa surface.[54] As in the Langmuir monolayers, it is possible for chemisorbed thiol131.4. Solid-Supported Bilayers and BiosensorsSAMs with more than one component to phase segregate into regions rich in one orthe other component, as shown by observing height differences in regions of thiolswith differing chain lengths by AFM in [55], for example.Self-assembly of physisorbed phospholipid bilayersThough Langmuir-Blodgett and Langmuir-Schaefer depositions provide a conve-nient means for depositing physisorbed layers onto a solid support, it is also possi-ble to self-assemble layers of certain molecules, especially phospholipid bilayers.There are several techniques available for self-assembling solid-supported bilay-ers. In one method, a freshly exposed metal surface is immersed in a solution ofthe lipid, then removed and immersed in an aqueous solution. [56] A more flexiblemethod involves immersing the substrate into a suspension of liposomes composedof the desired lipid and allowing them to burst onto the surface, forming a bilayer.[57] The exact mechanism for the formation of these bilayers from vesicles remainsunder investigation, but a number of theories have been proposed, as summarizedin Figure 1.11. Though both hydrophobic and hydrophilic substrates can be used,the mechanisms of formation differ greatly. On hydrophilic surfaces, the final bi-layer formed is cushioned by a 10-20 ? water layer near the support surface, thathydrophobic substrates necessarily lack. Adsorption of vesicles onto hydrophilicsubstrates begins with intact vesicles on the surface, with fusing and then rupturingto form a bilayer [59, 60]. Fusion of the liposomes is dependent on many factors,including surface chemistry, temperature, and osmotic pressure. [61] Simulationssupport this theory, indicating that the vesicle-surface interaction and membranetension are driving forces in vesicle bursting and bilayer formation. [62] The mem-brane tension of a vesicle is directly proportional to its radius, following Laplace?sLaw[63], and so vesicles of different sizes may be expected to have correspond-ingly different fusion behaviour. On some substrates, adsorbed intact vesicles willnot naturally form a bilayer even at high vesicle concentrations and other stimuli,such as freezing, must be employed in order to induce vesicle bursting. [64]Observations of this process for unilamellar DMPC (1,2-dimyristoyl-sn- glycero-3-phosphatidylcholine) vesicles by STM [65] showed that on a slightly hydropho-bic Au(111) substrate, the process of vesicle deposition and fusion proceeds byvesicle rupture to first form a monolayer that transforms in to a hemimicellar state,then to a complete bilayer as more vesicles burst onto the surface. Studies ofcholesterol-containing DMPC vesicles showed a similar behaviour. [66] On more141.4. Solid-Supported Bilayers and BiosensorsFigure 1.11: Proposed mechanisms of vesicle bursting and fusion to form supportedlipid bilayers (SLB). Reprinted from [58] with permission from Elsevier.151.4. Solid-Supported Bilayers and Biosensorshydrophobic surfaces, such as thiol SAM-modified substrates and uncharged mer-cury electrodes, liposomes will burst to form a monolayer at the interface. [67?69]This process appears to follow a similar two-step process to the deposition ontogold - fluorescence microscopy studies of giant unilamellar vesicles bursting onto asurface modified with a hydrophobic SAM showed that the vesicles first underwent?hemifusion?, creating an island of material from the outer leaflet, then bursting toform a monolayer across the surface. [70] Why multiple modes of adsorption ex-ist for these substrates is not entirely clear, however there is some suggestion thatthe mechanism is similar to a nucleation event, and can be described by the samemodels. [71] In this proposed model, the liposomes are substantially not attractedto a hydrophobic surface, but may be ?activated? and interact with the surface. Apossible mode of activation previously proposed [72] was that some of the hy-drophobic ?tails? of the phospholipid molecules randomly become exposed on theaqueous side of the membrane, providing a point of interaction with the hydropho-bic surface. The interaction of vesicles with metal surfaces such as these is highlydependent on the surface potential and charge density. These relationships will bediscussed in detail in Section 1.5.3.An advantage of forming solid supported bilayers by vesicle deposition is thattransmembrane proteins can be easily incorporated into the vesicles at formation,and will create a solid supported bilayer containing these proteins. [57, 73] Othermethods, such as the Langmuir-Blodgett deposition, are much less suited to pro-tein incorporation, as at some point the protein will be exposed to air and denature.A further improvement to the solid supported bilayers can come from modifyingthe initial liposomes to contain polyethylene oxide oligomers conjugated to phos-phatidylethanolamine (PEG-PE) lipids [74]. The long polymer tails on these lipidsextend into the aqueous region just outside the membrane (as shown in Figure 1.12)and help improve the stability of the solid supported bilayer when removed fromthe aqueous environment into air. However, these PEG layers do not provide asubstantial distance between the solid support and the bilayer, so incorporation ofmembrane proteins can be difficult if the protein extends into the space betweenthe bilayer and the solid support as it can interact unfavorably with the surface,adhering or denaturing. [75]161.4. Solid-Supported Bilayers and BiosensorsFigure 1.12: Depiction of PEG-PE containing vesicle and bilayer. Adapted withpermission from [74]. Copyright (2005) American Chemical Society.Interaction with an already present layerVesicles in solution are not limited to interacting with bare surfaces - as has beendescribed above, they are capable of bursting onto thiol SAM-modified substrates[60, 70] - if these SAMs are carefully chosen to contain a percentage of constituentswith an attached moiety (such as cholesterol [76] or a transmembrane protein [77])that can intercalate into a lipid bilayer, then vesicles from solution can burst ontothe interface, forming a bilayer that is attached to the underlying SAM throughthe linker moieties. These ?tethered? bilayers have an advantage over bilayers de-posited directly in that they are more stable than the physisorbed bilayers, and thelength of the molecule forming the tethering SAM provides a space between thebilayer and the solid support into which transmembrane proteins can project, re-ducing denaturing of even bulky proteins. [78, 79]It is also possible for liposomes in solution to incorporate into a solid supportedbilayer previously formed on a solid substrate. In the presence of proteins thatmediate membrane fusion, liposomes of egg phosphatidylcholine in solution wereobserved by fluorescence to incorporate into a solid supported bilayer of the samelipid. [80] The ability of liposomes to incorporate into such layers may not bea complete surprise, as incorporation of similar vesicles into cell membranes byendocytosis is common, however the observation of the process in vitro is worthyof note.171.5. Electrochemical BackgroundCreation of bilayers with templated patternsCreation of solid supported bilayers that encompass the entire substrate area is eas-ily achieved by the methods described above. However, it is not always efficient tohave an entire surface modified with one type of bilayer; individually addressableregions of bilayer with different lipid composition or a variety of incorporated pro-teins may be desired. In these cases, there are several methods in development forcreating surfaces that have spatially separated regions of lipid bilayer distributedacross the substrate. Diffusion barriers (Figure 1.13a) formed by standard photo-lithographic techniques can be used to create individual islands of bilayer on thesurface that will not mix [81, 82], as confirmed by fluorescence microscopy of thesurface. Similarly, microwells may be created on a surface, and different materialsintroduced into each well by micropipette. [83, 84] Microcontact printing [85] us-ing a stamp formed of a compatible material such as polydimethylsilane (PDMS)can be used to remove bilayer material in a pattern (?blotting?) or deposit bilayerdomains onto a clean substrate (?stamping?, Figure 1.13b). Other techniques, suchas nanolithography [86], oxidative removal of regions by UV illumination with aphotomask [87], and polymer stripping [88] have been demonstrated.Although many parameters of surface chemistry and geometry may be manip-ulated to influence the formation of solid supported bilayers, when using a conduc-tive substrate one of the easiest to manipulate is the electrical potential (and thusthe charge density at the interface). The electrode potential may be manipulated toencourage or discourage adsorption, or alter the structure of the adsorbed bilayer.These phenomena will be discussed in more detail in Section 1.5.1, but first anintroduction to the electrochemical background is warranted.1.5 Electrochemical Background1.5.1 TheoryThe behaviour of solvent and electrolyte molecules and ions near a charged surfaceis important for understanding electrochemical processes. At a molecular level,the solution side of the metal|solution interface may be imagined as consisting ofa mixture of solvent, anions, and cations, which are able to approach and interactwith the electrode surface in response to changes in electrical potential. Accord-ing to the Gouy-Chapman-Stern model of the interface [89], there are two distinct181.5. Electrochemical Background(a) Lithographically formed diffusionbarriers. From [81]. Reprinted with per-mission from AAAS.(b) Patterning by microcontact stamping.(c) Dip-pen nanolithography oflipid bilayers.Figure 1.13: Examples of different methods of forming bilayers on templated sur-faces.191.5. Electrochemical BackgroundFigure 1.14: Schematic model of the electrode-solution interface. Inner HelmholtzPlane (IHP) and Outer Helmholtz Plane (OHP) are shown along with potential (? )and charge density (? ). The hydration energy for anions is smaller than for cationsand therefore a solvation shell is not shown.regions near the electrode surface. Closest to the electrode is a monolayer of sol-vent molecules and any specifically adsorbed species (ionic or molecular), shownin Figure 1.14. The electrical centre of this layer is the Inner Helmholtz Plane(IHP), at a distance x1 from the electrode surface. Solvated ions cannot approachas close to the electrode surface; their closest approach is at distance x2, namedthe Outer Helmholtz Plane (OHP). Because of thermal motion, the entirety of thesurface charge on the electrode will not be compensated by ions within the OHP,but rather the charge compensation by electrolyte ions extends into the solution asthe diffuse layer for a distance (typically less than 100 ?). Since the interface isdescribed in this way as having two layers of charge distribution, the solution sideof the interface is often referred to as the ?double layer?.This model of the metal|solution interface suggests a behaviour similar to asimple capacitor conceptualized as two parallel plates, one the electrode surface,and the other the compensating charges in solution, beginning at the OHP andencompassing the diffuse layer. The capacitance of a parallel-plate capacitor may201.5. Electrochemical BackgroundFigure 1.15: RC Circuitbe generally described by the equationC = ??0d (1.3)where ? represents the relative dielectric constant of the material between the plates,?0 the permittivity of free space, and d the distance between the two plates. In theabsence of specifically adsorbed ions, the solvent and uncharged adsorbates at theIHP appear similar to the dielectric material. In fact, this model explains verywell electrode behaviour in solution, and the interface is typically modeled as anRC circuit - a resistor and capacitor in series, as highlighted in Figure 1.15, rep-resenting the so-called double-layer capacitance and the solution resistance. Thedouble-layer capacitance differs from a simple electrical capacitance in that it isoften potential-dependent due to changes occurring at the metal|solution interfacesuch as changes in double-layer structure for low electrolyte concentrations, surfacereconstruction of crystal faces [90?92], or adsorption/desorption events (see belowfor description). The presence of an electric field at the interface creates an electro-static pressure across the double-layer, compressing the solvent molecules presentat the IHP and altering their effective dielectric constant. [93] The double-layercapacitance is thus an important tool for monitoring the state of the metal|solutioninterface, and is often calculated experimentally by applying a small AC perturba-tion to the electrode and measuring the differential capacitance:C =(??M?E)?(1.4)where ?M represents the charge density of the electrode, E the electrode potential,and ? chemical potential.Adsorption of substances onto an electrode surface can alter the capacitance bydisplacing the solvent (taken here to be water) at the IHP with material with differ-ent dielectric properties. An electrode surface with adsorbed material at the surface211.5. Electrochemical Backgroundmay successfully be modeled as two parallel capacitors [94], one representing thecapacitance of the unmodified surface (C?=0) and the other the adsorbate-modifiedarea (C?=1), where ? is the fraction of surface coverage. The total capacitance isthus an area weighted average of the two capacitances. Equation 1.5 illustrates thisconcept, using ? as the fraction of electrode area covered by the adsorbate.Ctotal = (?)C?=1 +(1??)C?=0 (1.5)This model may also be used to estimate the fraction of surface covered by the ad-sorbate for systems where the adsorbate capacitance and bare-electrode capacitanceare known.Adsorbates may be desorbed or encouraged to adsorb onto an electrode surfaceby application of an electrical potential, because of changes in the interfacial ten-sion at the electrode surface. For the adsorption of a neutral adsorbate L onto anelectrode surface in a generalized electrolyte M+X? at constant temperature andpressure, the relation of the surface tension to electrode potential is given by theelectrocapillary equation:?d? = ?MdE++??d?MX +?Ld?L (1.6)In this equation, ? is the interfacial tension, ?? and ?L are the Gibbs surface excess(relative to water) of the X? ion and the adsorbate L, respectively, and ?MX and ?Lthe chemical potentials of the electrolyte and the adsorbate, respectively. A typicalelectrocapillary curve will be an inverted parabola when plotted versus electrodepotential, as shown in Figure 1.16 for two hypothetical surface states. As the elec-trode potential is changed the system will react to minimize the surface energy, sofor a given potential the stable state will be the one with the lowest interfacial ten-sion. In the example system of Figure 1.16, this would imply that the ? -state willbe the most stable only in a narrow potential window, and outside that region, the?-state will be most stable. This behaviour is demonstrated for the adsorption ofcamphor onto a mercury electrode [96] in Figure 1.17. Here, the camphor-modifiedelectrode (analogous to phase ? in Figure 1.16) is the more stable state betweenapproximately 0.4 and 1.8 V, and the unmodified electrolyte-electrode interface(phase ?) is otherwise more stable. The presence of camphor on the electrodesurface is indicated by the sharp decrease in capacitance in this potential range,caused by the displacement of electrolyte at the metal|solution interface with cam-221.5. Electrochemical BackgroundFigure 1.16: Effect of the increase of bulk activity of the surfactant (a: lowest, d:highest) on the interfacial tension ? , charge density ? and differential capacity Cfor two defined states ? and ? . The lowering of the interfacial tension (arrows in a)is larger for the most compact film. ? . Reprinted from [95] with permission fromElsevier.phor, which has a lower relative dielectric constant. Similar observations have beenmade for a range of adsorbates, among them isoquinoline [97], tert-pentanol [98],n-butanol [99], octyl alcohol [100], and lipids [101].Also notable in Figure 1.17 are the large, sharp peaks associated with the tran-sition between desorbed and adsorbed states. These so-called pseudo-capacitancepeaks are characteristic of moving between states on the metal|solution interface,due to kinetic limitations of the phase transition. In systems where there is a changein coverage with potential, current will be required to flow to the interface in or-der to accommodate the changing dielectric as coverage changes. The change incharge density with potential depends on the surface excess and the charge trans-ferred as the adsorbate is transferred on or off the surface. Although this responseis complex, Equation 1.4 can be modified to account for the charge transfer (andthus change in charge density) to become:C =(??M?E)?=(??M?E)?+(??M??)E(???E)?(1.7)The second term in this new equation will be dependent on the AC frequency used,231.5. Electrochemical BackgroundFigure 1.17: Differential capacitance-potential curve of a HMDE in: (1) a 0.5 MNa2SO4, (2) 0.5 M Na2SO4 solution saturated with camphor. Frequency: 45 Hz.Reprinted from [96] with permission from Elsevier.241.5. Electrochemical Backgroundas the change in coverage with potential((???E)?)will vary depending on thekinetics of the phase transition. In other words, as the rate of the phase changedecreases, the ability of the system to respond to the higher frequency AC pertur-bations decreases. This effect is shown for the adsorption of octyl alcohol in Figure1.18. Note that as the AC frequency used to measure the differential capacitanceincreases, the height of the pseudo-capacitance peaks decreases. At the limit ofinfinite frequency, the peaks disappear entirely and the transition between C? andC? is smooth. At the limit of zero frequency, the expression reduces to Equation1.4.The frequency of any applied AC perturbation will have an additional effect onsolid electrodes, due to the special nature of their interfaces. Due to microscopic oratomic-scale roughness [103], they may exhibit a distribution of relaxation times,rather than the single relaxation time implied by the simple RC circuit model gen-erally used in capacitance calculations. In this case, a constant-phase element [104]resembling an infinite series of parallel RC circuits may be used rather than a ca-pacitor when modeling the cell. Despite this, it has been shown [105] for polishedAu(111) electrodes that the simpler RC circuit is a good approximation of the in-terface behaviour, as little frequency dispersion in the range 1 kHz to 0.1 kHz wasfound in the absence of specifically adsorbed material.1.5.2 Octadecanol ElectrochemistryOctadecanol (C18H37OH) as a thin film has been studied extensively. Its character-istics as a monolayer are well known, as summarized in Section 1.3. As such, ithas lent itself well to electrochemical studies of its properties also. As a floatingmonolayer, it has already been noted that at pressures close to the ESP there are de-fects in the layer such that oxygen is able to diffuse from the air into the subphasebelow. [39] As will be described below, the presence and characteristics of thesedefects will dominate the electrochemical behaviour of octadecanol adsorbed ontoan electrode surface.Monolayers of octadecanol on gold and mercury electrodesAlthough essentially similar to the soluble surfactants described in Section 1.5.1,the electrochemical behaviour of electrodes coated with a monolayer of octade-canol has some distinguishing properties, mostly linked to its insolubility in aque-251.5. Electrochemical BackgroundFigure 1.18: Differential capacity of a Hg electrode in 1 M KNO3. Solid linesshow the capacitance of the electrode in a saturated octyl alcohol solution at 0.24and 10 kHz. Heights of the pseudocapacitance peaks at intermediate frequenciesare marked with arrows. The dashed line shows the capacitance of the electrode inan octyl alcohol-free solution. Reprinted from [102], Copyright Wiley-VCH VerlagGmbH & Co. KGaA. Reproduced with permission.261.5. Electrochemical BackgroundFigure 1.19: Capacitance behaviour of octadecanol on a HMDE. Potential isgiven versus the pzc of the interface. Dotted line: 0.05 M KClO4. Solid line:0.05 M KClO4 with a monolayer of octadecanol deposited from the gas|solutioninterface. Reproduced from [107], Copyright Elsevier 1997.ous electrolyte. For a monolayer deposited onto a mercury electrode by gentlypushing the electrode through an octadecanol-coated gas|solution interface [106],the capacitance behaviour has similar general characteristics to that of the previ-ously described for camphor. (Camphor: Figure 1.17, octadecanol: Figure 1.19.)In the potential range at which the adsorption of the octadecanol is favoured, fromapproximately 300 mV measured against the potential of zero charge (pzc) to ap-proximately -350 mV/pzc, the capacitance is significantly lowered as comparedto measurements in the absence of octadecanol, much as for the camphor pre-viously described. In contrast with the camphor example, however, the pseudo-capacitance peaks characteristic of adsorption/desorption events are muted in theoctadecanol system. This difference is believed to be due to the slow kinetics ofadsorption/desorption of octadecanol, which is evident also in the shift in potentialof readsorption vs. desorption. In mercury, the minimum capacitance is found nearthe pzc, and was measured to be approximately 10 ?F/cm2.On a single-crystal Au(111) electrode, the positive desorption potentials are notaccessible due to an irreversible oxidization of the adsorbate, [106] but the negativedesorption limit is accessible. A typical differential capacitance measurement ofthe desorption and readsorption of an octadecanol monolayer is shown in Figure1.20. The minimum capacitance (again near the pzc, which is 0.330 V/SCE forAu(111) in 0.1 M NaF [108]) is similar to that on Hg, indicating that the mono-271.5. Electrochemical Backgroundlayer is not disturbed by the deposition onto a solid surface, which will necessarilyhave some roughness, rather than the smooth mercury surface. Owing to the morecomplex phase transition involved in the desorption from Au(111), the number andheight of the pseudocapacitance peaks is different, but the hysteresis between des-orption and readsorption potential remains. The position of these peaks (and thusthe associated desorption/readsorption events) was found to depend on the elec-trolyte concentration, [109] shifting the peaks approximately -100 mV as the con-centration was varied from 100 mM to 5 mM. Therefore, the transition is driven bychanges in interfacial tension.The quality of the layer, as measured by the number and height of pseudoca-pacitance peaks, also strongly depends on the surface pressure of the octadecanolmonolayer before it is transferred to the electrode surface. [110] As shown in Fig-ure 1.21, at the lowest film pressures the minimum capacitance is greater, whichis expected given the lower molecular density of the layer and higher number ofdefects in the less condensed phases (as also measured in [39]). Studies of mixedmonolayers of octadecanol and oleyl alcohol, miscible with each other, showeda decrease in the minimum capacitance as the mole fraction of oleyl alcohol inthe monolayer was increased. [111] Taken with the findings on film pressure, thissupports the conclusion that octadecanol monolayers, especially those depositedat moderate film pressures (i.e. at the ESP) contain defects at deposition that dotransfer to the electrode surface. Comparing theoretical and actual differential ca-pacitance measurements of octadecanol monolayers deposited on Au(111) by bothLangmuir-Schaefer and Langmuir-Blodgett methods suggest that the best surfacecoverage is about 0.9. [112] As the initial film pressure increases, the pseudo-capacitance peaks at desorption become gradually smaller and broader, and shifttoward more negative potentials, exaggerating the hysteresis in potential at desorp-tion/readsorption (not shown in Figure 1.21). These investigations, supported bystudies of mixed monolayers of the immiscible pyrenenonanol/octadecanol mixture[111] highlight that the state of the monolayer at the gas|solution interface beforetransfer to the electrode is a major factor in the state of the adsorbed monolayer.Although the pseudocapacitance peaks inform us that there is a phase changeoccurring, as a transition from one state?s electrocapillary curve to another, theydo not provide any information about the nature of this change. Investigationson the film pressure changes of Au(111) modified with an octadecanol monolayer[114] showed a fit to three electrocapillary curves with transitions occurring at -281.5. Electrochemical BackgroundFigure 1.20: Cyclic voltammogram (a) and differential capacitance (b - dashed line:negative scan. solid line: positive scan) of an octadecanol monolayer on Au(111) in0.05 M KClO4. Dotted lines show CV and capacitance of electrode in the absenceof octadecanol. Reproduced from [113]. Copyright Elsevier 2004.Figure 1.21: Effect of varying initial surface pressure on differential capacitancescans. Scans are negative-going desorption scans of an octadecanol monolayer onAu(111). Reprinted from [110] with permission from Elsevier.291.5. Electrochemical BackgroundFigure 1.22: Elastically scattered light on octadecanol monolayer Au(111). Repro-duced from [113]. Copyright Elsevier 2004.250 mV/SCE and -600 mV/SCE, implying that three states are necessary to de-scribe the desorption process. Adding another form of measurement in-situ, suchas fluorescence (discussed later in Chapter 5), infrared spectroscopy, or elasticallyscattered light measurements, can help determine the processes involved in thetransition. Light scattering studies done on octadecanol monolayers adsorbed onAu(111) [106, 113] (Figure 1.22) show constant scattering at potentials more posi-tive than -600 mV/SCE, and a sharp increase in scattering negative of -600 mV/SCEas the potential is scanned negatively. This behaviour suggests that the octadecanollayer remains very near the electrode surface at the more positive potentials, thenmoves away from the surface as some form of small aggregate causing scatter-ing at -600 mV/SCE. These particles or aggregates must remain relatively closeto the electrode surface to remain within the focal volume of the measurement.On the positive readsorption scan, the scattering stays high until approximately -300 mV/SCE, matching the onset of the first change in capacitance, indicating thatthe octadecanol does remain desorbed from the surface. The hysteresis in the lightscattering measurements matches with those in the electrochemical measurements,supporting the hypothesis that the octadecanol desorbs from the electrode surface,but remains near, and that a difference in mechanism of desorption vs. readsorp-tion must contribute to the potential shift between these two processes. Polarizationmodulation infrared reflection absorption spectroscopy (PM-IRRAS) of the octade-canol monolayer on Au(111) [112] also show an increase in the octadecanol tiltangle that loosely correlates with the changes in capacitance (Figure 1.23).301.5. Electrochemical BackgroundFigure 1.23: Changes in octadecanol chain tilt angle as determined by PM-IRRAS.Squares represent the first negative scan; circles the first positive scan. Solid line isthe differential capacitance trace of the negative scan. Reproduced from [112] withpermission from Elsevier.Bilayers of octadecanol on gold electrodesFewer studies of octadecanol bilayers have been undertaken than of the monolayers.However, the electrochemistry is well-characterized. The capacitance behaviour(Figure 1.24) shows a pattern of behaviour related to the monolayer capacitance.The capacitance is initially lower than a monolayer, as expected since not only isthe dielectric octadecanol layer thicker, but most defects present in a monolayerare covered over by the second layer, better isolating the electrode surface. Atapproximately -200 mV/SCE, a small pseudocapacitance peak is associated withan increase in capacitance of the layer, indicating a phase change in the layer. Thishas been hypothesized to be a formation of defects in the adsorbed layer, as depictedin the cartoon (e) of Figure 1.24. At approximately -600 mV/SCE, a large increasein capacitance corresponds with the desorption of the bilayer from the electrodesurface (Figure 1.24a). Elastically scattered light measurements [106] indicate that,as for the monolayers, desorption results in small aggregates that remain near theelectrode surface. The mechanism of readsorption of these presumed aggregatesmust be different from the desorption, as indicated by the large hysteresis in thecapacitance scan. One possible mechanism (Figure 1.24b-c) is that the aggregatesfirst form a defective monolayer, then undergo a phase transition (indicated by the311.5. Electrochemical BackgroundFigure 1.24: Proposed mechanism of desorption and readsorption of a bilayer ofoctadecanol on Au(111). Reprinted with permission from [115]. Copyright 1999American Chemical Society.pseudocapacitance peak at approximately -150 mV/SCE) to re-form the adsorbedbilayer.The structure of the adsorbed octadecanol bilayer on Au(111) has been furtherinvestigated by PM-IRRAS [46]. Layers deposited by two sequential Langmuir-Schaefer touches, as well as X- and Y-type Langmuir-Blodgett depositions, werecompared electrochemically and spectroscopically. Although all three types of de-positions formed layers with some basic similarities in their electrochemical be-haviour (e.g. low initial capacitance, desorption at negative potentials, hysteresisbetween desorption and readsorption potentials), both the Langmuir-Schaefer andX-type layers showed an irreversible change in behaviour after the first desorptionscan, becoming more like the Y-type layers after repeated scans, while the Y-typelayers remained stable, as seen in Figure 1.25. Based on this behaviour and thechanges in carbon-chain angle measured by PM-IRRAS, it was determined that theX-type and Langmuir-Schaefer bilayers, which initially have a head-to-tail configu-ration (see Figure 1.10) rearrange upon desorption to form a more kinetically stableY-type configuration in the readsorbed layer. This transformation begins after a sin-gle desorption, although it takes several cycles to become complete. Although the321.5. Electrochemical Background(a) Langmuir-Schaefer bilayer (b) X-type bilayer (c) Y-type bilayerFigure 1.25: Differential capacitance of Langmuir-Schaefer, X-type, and Y-typebilayers of octadecanol on Au(111). Trace 0: clean Au(111). Trace 1: Initial des-orption (negative-going) scan. Trace 2: Initial readsorption (positive-going) scan.Trace 3: Equilibrium desorption & adsorption scans. Reprinted with permissionfrom [46]. Copyright 2004 American Chemical Society.X-type layers are stable while adsorbed, the conversion to Y-type is irreversible.This state change has important implications for layers formed by the Langmuir-Schaefer touch method, as the as-deposited layer will not be directly comparable toa layer that has experienced a desorption-readsorption cycle.1.5.3 Electrochemistry of DOPC and VesiclesPhospholipid bilayers and monolayers may be deposited directly onto an electrodesurface, either by deposition from a floating layer or from vesicles in solution. Un-derstanding the electrochemical behaviour of such layers is essential to designingfunctional electrochemical biosensors.Monolayers of phospholipid on mercury electrodesThe electrochemical study of monolayers was quickly identified as an importantpathway for characterizing bilayer behaviour at a surface. Monolayers of phos-pholipids deposited onto a HMDE were first studied by Miller et. al. [116?119],including studies on the incorporation of lipoproteins into the monolayer and oxy-gen transport across the interface. Monolayers of synthetic lecithin (a mixture ofphosphatidylcholines) show a similar differential capacitance behaviour to otheradsorbed surfactants on mercury electrodes, viz. a ?capacitive pit? at potentialswhere it is readily adsorbed onto the electrode, with pseudocapacitive peaks mark-ing the departure from the adsorbed state into desorbed or alternate phases. For pure331.5. Electrochemical BackgroundFigure 1.26: Differential capacitance of L-?-lecithin (open circles) and DL-?-lecithin (closed circles) monolayers on a mercury drop electrode. Potential mea-sured versus normal calomel electrode (NHE). Note that the x-axis is reversed fromother figures in this chapter. Reprinted from [116] with permission from Elsevier.phospholipids, the capacitive behaviour is well-defined, as in the case of DOPC,where the peaks visible at approximately -1 V/SCE become sharper (Figure 1.27),and are well-characterized enough that they are used to characterize the monolayerbehaviour in the presence of membrane-altering additives, such as antibiotics [120]and hydrophobic hydrocarbons [121, 122]. The first peak is believed to be due toa phase change creating defects in the layer [123] and the second, smaller, peak toa nucleation and growth process of these defects. [124] At more negative poten-tials, the layer is desorbed, although it remains in close proximity to the electrodesurface.Impedance spectroscopy studies of the mercury-supported DOPC monolayersuggested that the simple model of an RC circuit for the electrode-electrolyte inter-face, as described in Section 1.5.1, may not be adequate for describing the modifiedHg interface. [67] In this study, the solution resistance was found to increase dra-matically when the DOPC monolayer was present on the mercury surface.341.5. Electrochemical BackgroundFigure 1.27: Differential capacitance of a DOPC monolayer adsorbed from thegas|solution interface on a mercury drop electrode. Reprinted from [125].Potential-dependent behaviour of phospholipid bilayers on electrode surfacesPhospholipid bilayers are only stable on Au(111) in a limited range of potentials.Studies of DOPC and DMPC bilayers formed from vesicles in solution show thatthe layer is stably adsorbed from ~0.2 V/SCE to ~-0.35 V/SCE, negative of whichthere is some sort of phase transition, followed by the desorption of the layer fromthe surface at ~-1.1 V/SCE. Both DOPC and DMPC have similar capacitance be-haviour, although the transitions for DOPC are less well defined than for DMPC(Figure 1.28). For supported bilayers at room temperature (20 ?C), DMPC is inthe ripple state, and DOPC is in the liquid crystalline state, as determined by PM-IRRAS. [126] Liposomes of DOPC, when adsorbing from a suspension, adsorbwithout bursting for electrode charge densities less than 8 ?C/cm2, but at greatercharge densities forms a bilayer (or multilayer), as determined by EQCM measure-ments. [127]Impedance and chronocoulometric measurements of the DOPC monolayer onmercury drop electrodes have shown [123, 128] that at potentials more negativethan -0.65 V/SCE, the monolayer is stable and is defect-free, until the first phasetransition at approximately -0.925 V/SCE, and the monolayers are displaced withapplication of potentials below -1.8 V/SCE.DOPC, when adsorbed onto a gold surface, is less mobile at positive chargesthan at negatively charged gold. This is likely due to Coloumbic interactions be-tween the charged headgroup and the electrode surface, likely an increased local-ization of the negative charge on the phosphate group. [127]351.5. Electrochemical BackgroundFigure 1.28: Differential capacitance desorption scans of DOPC (dotted line) andDMPC (solid line) bilayers formed from vesicles in solution. Reprinted from [126]with permission from Elsevier.In-situ neutron reflectivity studies of a DMPC bilayer formed by vesicle depo-sition onto an Au surface show that the layer is thinner at more positive potentialsthan the bilayer thickness given by X-ray diffraction, suggesting that the tilt angleof the DMPC tails is larger. As well, at these more positive potentials, the bi-layer incorporates more of the water solvent into the bilayer structure (as depictedin the cartoon schematic of Figure 1.29. As the electrode potential is moved tomore negative values, the layer becomes thicker and eventually desorbs from theelectrode surface, although it stays quite nearby, even at -950 mV/SCE, being sep-arated only by a thin layer of the water solvent that is up to approximately 10 ?thick. This ?desorbed? membrane closely resembles the structure of bilayers sup-ported on quartz substrates. Here the bilayer is known to be separated from thesubstrate by a thin layer of solvent. Charge density measurements show that thebilayer is stable on the electrode surface at charge densities of an absolute valueless than 8 ?C/cm2. Outside this range, the charge density curve matches with thatof the layer-free electrode surface. This behaviour matches with that observed byneutron reflectivity. [129, 130]Characterization of DMPE vesicle spreading under potential control suggeststhat the spreading onto the vesicles is slow, based on hysteresis between positiveand negative differential capacitance scan directions. Spreading of vesicles takesplace between -600 and -400 mV (positive scan direction), and forms a condensedbilayer phase above -400 mV. However, this layer contains many defects and is notas condensed as layers that form on Hg, as characterized by capacitance and film361.5. Electrochemical BackgroundFigure 1.29: Cartoon representation of the behaviour of DMPC on an Au electrodeunder potential control. Blue dots represent water, red dots the hydrophilic head ofthe DMPC lipid. Reprinted from [129] with permission from Elsevier.371.5. Electrochemical Backgroundpressure. Based on capacitance, the coverage for films formed in this way is ap-proximately 80%. PM-IRRAS spectroscopy confirms that the tilt angle of the acylchains of DMPC is large when adsorbed onto the electrode surface (approximately55 degrees with respect to the surface normal). [131] The orientation of the polarheads in the two leaflets may differ due to their exposure to different environments.[132]Further studies of bilayers of DMPC on Au(111) provide an estimate of surfacecoverage to be ~75% based on differential capacity measurements, and confirm thelarge tilt angle and increased head group spacing due to hydration in the adsorbedstate. [133] Addition of cholesterol or other molecules that increase layer stiffnessmay be needed to achieve the lowest capacitances in a solid supported lipid bilayer.[134] Fatty alcohols such as octadecanol are soluble in the bilayer and may also beused to similar effect. [135]Vesicle behaviour in an electric fieldAdsorption of vesicles of DOPC on polycrystalline gold (sputtered on quartz QCMsubstrate) depends on the charge density (and thus potential) of the electrode sur-face. At potentials near the pzc of the interface, data from EQCM suggest that 20nm vesicles of DOPC adsorb whole, rather than bursting onto the surface. At po-tentials positive or negative of the pzc, however, the data suggests that a bilayer-likesystem forms as vesicles interact with the electrode surface. [127]Adhesion of phospholipid vesicles can be observed by chronocoulometric mea-surements, pinpointing individual adhesion events. [136] Using multilamellar DOPCvesicles, the current spikes (shown in Figure 1.30) are unidirectional, with the direc-tion of current flow determined by the electrode charge density. The bidirectionalsignals are produced when the charge density on the electrode surface is less thanthat of the choline groups of the phospholipid at the vesicle-electrode interface,and were observed as well in unilamellar DOPC vesicles and vesicles of DMPCand DPPC. The bidirectional shape results from the initial current flow due to li-posome contact with the electrode, displacing ions (negative portion) followed bythe interaction of the positively charged choline groups with the negatively chargedelectrode (positive portion). The bidirectional peak is only visible for potentialswhere the charge density is less than the charge density created by the cholinegroups.This same technique has previously been used to measure the adhesion of cells381.5. Electrochemical Background(a) Unidirectional signals (b) Bidirectional signals.Figure 1.30: Adhesion signal for multilamellar DOPC vesicles adhering to a mer-cury electrode. Reprinted from [136] with kind permission from Springer Scienceand Business Media.and oil droplets onto mercury electrodes. [137?141]Liposomes of DOPC have been shown to interact in a potential-dependent man-ner with a monolayer of DOPC existing on a mercury electrode. [124] The mono-layer formed by either deposition from the gas-solution interface or by liposomaldeposition were initially similar (Figure 1.31, light traces), however upon excursionto potentials more negative than the second capacitance peak, liposomes in solutionare able to interact with the electrode through the newly formed defects. The result-ing layer has properties substantially different than the initially formed monolayer,most noticeably the two capacitance peaks are reduced in size and shifted towardsmore positive potentials. This change in behaviour remains even when the poten-tial is scanned past the desorption potential of the monolayer. The cartoon in Fig-ure 1.32 proposes a mechanism for the interaction of liposomes with the adsorbedmonolayer during these potential excursions.Free vesicles in solution may also be affected by the electric field induced bythe electrode potential; in studies of giant unilamellar vesicles (larger than 1 ?m di-ameter), exposure to alternating electrical fields caused an elongation of the vesicleinto an oval shape. [142] Compared to the electrochemical behaviour previouslydiscussed for vesicles, these fields are large (ca. 20 V RMS) and higher frequency(ca. 1-15 kHz). [143] Exposure to lower-frequency or DC fields can induce elec-troporation and rupture of vesicles in solution, and has been explored as a methodof drug or gene delivery. The potential and frequency requirements for electropo-ration are dependent on several factors, such as vesicle size and distance from the391.5. Electrochemical BackgroundFigure 1.31: Differential capacitance of DOPC on a mercury electrode. (a)Capacitance-time profile of liposome adsorption. (b) DOPC adsorbed solely fromthe gas-solution interface. (c) DOPC adsorbed from liposomes in solution. (d)DOPC adsorbed from the gas-solution interface with liposomes in solution. Lightlines: initial (negative) scan. Heavy lines: positive scan. Reprinted from [124].Copyright Elsevier 2001.401.5. Electrochemical BackgroundFigure 1.32: Proposed mechanism for liposome interaction with an existing mono-layer of DOPC. Inset: Formation of a monolayer from liposome solution. Repro-duced from [124]. Copyright Elsevier 2001411.6. Fluorescence Backgroundelectrode, however conditions that provide a transmembrane voltage of more than54 mV have been shown to be sufficient to induce pore formation in DOPC vesi-cles. [144] For giant vesicles (7.5 ?m diameter), this corresponds to a 5.0 V RMSpotential applied at frequencies between 20 Hz to 1 kHz in a 0.1 mM NaHCO3buffer. [145] An increase in electrolyte strength would be expected to increase thevoltage required for electroporation due to the shielding effect of electrolyte ionsbetween the vesicle and electrode (Section 1.5.1).1.6 Fluorescence Background1.6.1 TheoryMolecular fluorescencePhotons of light in the UV or visible range may interact with molecules by ab-sorbance, creating an electronic excited state in the molecule. Obeying the quan-tum mechanical selection rule ?S = 0, these transitions will result in an excitedstate of the same electronic multiplicity (usually singlet) as the initial electronicstate. Transitions directly between states of differing multiplicities are forbidden,and thus rare, for an absorption event. Once the molecule is in an electronic ex-cited state, however, it is possible to reach a state of a different multiplicity throughintersystem crossing. Relaxation of the excited state can occur by any of severalpathways, as summarized in Figure 1.33. Most commonly, the molecule can re-lax non-radiatively, by vibrational transfer of energy to other nearby molecules.This may be a partial relaxation within the excited state, or may release enough en-ergy to return the molecule to the ground state entirely. From the excited state, themolecule may also relax by release of a photon (fluorescence). As some vibrationalrelaxation generally occurs before release of a fluorescent photon, it is typically ofa slightly lower energy than the absorbed photon. This energy loss is typical ofmolecular luminescence phenomena, and is called the Stokes shift. Phosphores-cence may also occur if the molecule has a triplet state accessible from the excitedstate.There are many other possible fates for a molecule after achieving an electronicexcited state. The excited molecule may react with another molecule in the groundstate, forming an excimer. The molecule may also undergo photo-bleaching, apermanent loss of fluorescence due to a breakdown of the fluorescent molecule421.6. Fluorescence BackgroundFigure 1.33: Partial energy diagram for a photoluminescent system.itself, or be quenched by energy donation to other species (a non-permanent loss offluorescence). Quenching has many causes, from collisional quenching by oxygenor heavy atoms to energy donation to other molecules or even a metal surface,discussed in more detail below.F?rster resonance energy transfer (FRET)One decay pathway of an excited fluorophore is to lose its energy through non-radiative transfer to another molecule. In the case of resonance energy transfer, theenergy is transferred by long range dipole-dipole interactions between the energydonor and acceptor molecules. The acceptor molecule may relax radiatively ornon-radiatively after the transfer. The rate of this transfer , kT , is expressed askT (r) =1?D(R0r)6(1.8)where ?D is the decay time of the donor molecule in the absence of the acceptor, r isthe separation between the donor and acceptor, and R0 is the F?rster distance - the431.6. Fluorescence Backgrounddonor-acceptor separation at which the energy transfer is 50% efficient. The ratealso depends on factors such as the relative orientation of the dipoles, the quan-tum yield of the donor, and the spectral overlap between the donor and acceptormolecules, that are incorporated in to the F?rster distance value:R0 = 0.211(?2n?4QDJD(? ))1/6 (1.9)Here, the F?rster distance (in ?) is shown to depend on ? , describing the relativeorientations of the donor and acceptor?s transition dipoles, n, the refractive index ofthe medium between the donor and acceptor, JD, the overlap integral describing thedegree of spectral overlap between the emission of the donor and the absorption ofthe detector, and QD, the quantum yield of the donor in the absence of the detector.Typically, the F?rster distance is between 20 to 60 ?. [146]From Equation 1.8, it can be seen that the rate is strongly distance dependent,falling off with r?6. This sensitivity to donor-acceptor separation is exploited inseveral techniques, such as determination of protein folding structure [147?149],DNA and RNA binding (Figure 1.34) [150?153] , and lipid distribution in cellmembranes [154?157], among others.The BODIPY fluorophore (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene) is knownto form two dimers, termed DI and DII (Figure 1.35). Mikhalyov et al. [158], Tleu-gabulova et al. [159], Bergstr?m et al. [160] Both can accept energy through FRETfrom nearby monomers. The DI dimer is non-fluorescent and exhibits an absorp-tion band blue-shifted from the monomer, while the DII dimer has a red-shiftedabsorbance, and has an emission maximum at approximately 577 nm, red-shiftedcompared to the monomer emission maximum at approximately 505 nm. TheF?rster distance for the BODIPY dimers is relatively long, at 57 ? for the monomer- DI transfer, and 42 ? for the monomer - DII transfer. [158] This self-dimerizationhas been harnessed to monitor relative concentration of fluorophore in surfactantlayers, Musgrove et al. [1], Casanova-Moreno and Bizzotto [161, 162], Musgroveet al. [163], Shepherd et al. [164], protein folding, [160] membrane organization,[165] and light-harvesting antennae for artificial photosynthesis.[166]Fluorescence near a metal surfaceWhen the oscillating electric field of a fluorophore is near a metal surface, the near-field radiation from the fluorophore can interact with the metal surface in several441.6. Fluorescence BackgroundFigure 1.34: Application of FRET to RNA sensing. (A) Two labeled oligonu-cleotides hybridize to the target sequence, bringing the fluorophores close enoughtogether that RET is observed. (B) Changes in fluorescence spectrum observedwith hybridization. (Donor emission maximum: 503 nm, acceptor emission max-imum: 670 nm.) (a) Fluorescence spectrum of the unbound probes.(b) Spectrumin the presence of the target DNA sequence. (c) Spectrum in the presence of anontarget DNA sequence. Reprinted from [150] with permission from Elsevier.Figure 1.35: Proposed structures of the BODIPY dimers. Left: the DI dimer. Thedistance between the ring planes is 4.9 ?. Right: The DII dimer. The angle betweenthe transition dipoles is 55?, and the distance between centers-of-mass is 3.8 ?.451.6. Fluorescence BackgroundFigure 1.36: Excited state lifetime of Eu3+in air near a silver mirror surface. Dots:Experimental data. Solid line: best fit to data. Reprinted with permission from[168]. Copyright 1975, American Institute of Physics.Figure 1.37: Oscillating field of dipole interacting by interference with a planemirror constructively (left) and destructively (right). Reprinted from [167] withpermission from Elsevier.ways. At relatively long distances (greater than about 100 nm), the field can re-flect off the surface and interfere with the field radiating from the fluorophore. Thisresults in an oscillating pattern of enhanced and reduced fluorescence with separa-tion of the fluorophore from the surface, as established by early measurements ofeuropium fluorescence near a silver mirror, shown in Figure 1.37. [167]At shorter separations, the behaviour becomes less easily explained. In somecases, fluorescence near a metal surface can be enhanced [169]. Raman scatter-ing may be similarly enhanced, as seen in the surface-enhanced Raman scattering(SERS) technique. [170] Quenching also commonly occurs at very small separa-tions between fluorophore and metal surface, under similar conditions. The quench-461.6. Fluorescence Backgrounding follows a cubic relation with separation, as expected for a F?rster-type energytransfer between the fluorophore and surface. [171] The process of quenching isusually described as energy transfer into so-called ?lossy surface waves? in themetal.It was determined by Lakowicz [172, 173] that both enhancement and quench-ing processes are caused by coupling of the radiative field from the fluorophore intoplasmons in the nearby metal. In enhancement phenomena, the plasmons, locatedat the surface, are able to radiate the energy. However, in quenching the plasmonsare restricted by optical properties of the metal surface, are unable to radiate, andmust dissipate the energy as heat. The mode of energy transfer to a nearby metalsurface depends on the separation between the fluorophore and metal; at distancesgreater than 100 nm radiative decay of the excited state dominates, from 10 to400 nm, coupling with plasmons occurs, and below 10 nm, coupling into the lossysurface waves - plasmons that dissipate as heat - is dominant. The effect dependson the nature of the surface (including thickness of the metal) and on the geometry,with fluorescence enhancement being improved by surface roughness or corruga-tions on the order of the light wavelength. [174] This principle has been appliedextensively in the study of adsorbed organic films, discussed below, as well as inthe field of Total Internal Reflection Fluorescence (TIRF), where it allows enhancedimaging of structures just above a thin (less than 100 nm) metal-coated interface(eg. cytoplasmic and cellular regions) while quenching fluorescence from the por-tion of the membrane in direct contact with the interface.[175]The quenching phenomenon has also been exploited in making DNA-basedsensors (Figure 1.38). In these sensors, a SAM of fluorophore-tagged DNA isformed on an electrode surface. By application of an alternating potential, the con-formation of the DNA strands can be altered. As DNA has an inherent negativecharge at most pH values, it will be attracted towards the electrode surface at pos-itive potentials, bringing the fluorophore closer to the surface and quenching it. Atnegative potentials, the DNA is repelled from the surface, lifting the fluorophoreaway from the metal and reducing its quenching. Single-stranded DNA is muchmore flexible than double-stranded DNA, so changes in the frequency response ofthis system to potential perturbations can give information as to the binding stateof the DNA at the surface. This has been successfully employed to develop DNAsensors able to detect analyte concentrations of less than 10 pM (a surface densityof approximately 3?108interactions per square centimeter). [176, 177]471.6. Fluorescence BackgroundFigure 1.38: Fluorescence sensor for DNA binding. At negative potentials (top),the negatively charged DNA backbone is repelled from the surface and stands erect,dequenching the fluorophore. At positive potentials, the DNA is attracted to theelectrode surface and lies flat, quenching the fluorophore. Reprinted with permis-sion from [176]. Copyright 2004 American Chemical Society.Optics and microscopyFluorescence is conveniently measured by optical microscopy, providing not onlya general intensity measurement, but a spatial distribution of the fluorescence asan image. In order to facilitate in-situ imaging of thin films, either conventionalor episcopic fluorescence (epi-fluorescence) microscopy may be used. In an epi-fluorescence microscope, both the excitation light and the emitted light pass throughthe microscope objective. In order to restrict the light to the appropriate wave-lengths, a series of filters is used, as outlined in Figure 1.39a. Typically, holo-graphic filters are used with the sharpest cutoff possible in the wavelength rangeto minimize spectral overlap between the filters. Transmission profiles for a typ-ical filter set are shown in Figure 1.39b. Inverted microscope geometries, wherethe objective looks up on a sample from below, may also be used to accommodatea single-crystal electrode in a hanging meniscus configuration for in-situ imagingthrough liquid subphases.The choice of objective for epi-fluorescence microscopy is especially impor-tant, as it provides both the incident light and gathers the fluorescent light for anal-ysis. The function of the objective is limited by the design and composition of itscomponent lenses. Individual lenses will have some degree of spherical aberrationand spectral aberration, causing light of different wavelengths have slightly differ-481.6. Fluorescence Background(a) Schematic of the light path for an invertedepi-fluorescence microscope. The excitation fil-ter selects the desired band from the white lightsource, which is then reflected off of a dichroicmirror. Emitted light from the illuminated objectis captured by the objective, and passes throughthe dichroic mirror to the camera detector. Re-flected incident light is removed by the emissionfilter.(b) Typical filter profiles for (1) emission (2) ex-citation and (3) dichroic mirror filter cube compo-nents.Figure 1.39: Components of fluorescence focal points. By combining several lenses of different shapes and refractiveindices, this aberration can be minimized. Several objective designs are commer-cially available that correct spectral aberrations to a greater or lesser degree, butcurrently the best correction comes from ?apochromatic? lenses. Minimizing thechromatic aberration in fluorescence microscopy is not only important to obtaincrisp focus in images, but since the excitation and emission wavelengths beingused are distinct, ensuring that both share a focal plane is important for obtainingthe best signal in the fluorescence measurements.The resolving power and sensitivity of an objective is a function of it?s numer-ical aperture (NA). The NA of a lens is defined asNA = n? sin(?) (1.10)where n is the refractive index of the medium (typically air or water), and ? is thehalf-angle of the light cone (Figure 1.40). The numerical aperture thus affects thefocal length of an objective, and also its resolving power. A point source of light,upon passing through the series of optics in a microscope, will produce a diffrac-tion pattern called an ?Airy disc?. The Rayleigh criterion for defining the resolu-491.6. Fluorescence BackgroundFigure 1.40: Light cone of a microscope objective.tion limit of a microscope assembly is the separation between two point sources atwhich the central diffraction spot of one point coincides with the first diffractionminimum of the other point. This is depicted graphically in Figure 1.41. The the-oretical limit of spatial resolution for an objective is determined by the wavelengthused and the numerical aperture using the Abbe formula:d = 0.61?NA(1.11)where ? is the wavelength of light used, NA the numerical aperture as defined inEquation 1.10. The quantity d corresponds to the minimum radius of a particle thatcan be accurately imaged under these conditions. Features with a radius smallerthan d may be visible, but their size will be convolved to the size of the diffractiondisk and thus will have an apparent radius of d, regardless of their true size. [178]For an objective with NA=0.5 and a wavelength of 500 nm, the diffraction limitwould be 610 nm, however this resolution is rarely reached in practice. Severaltechniques may be applied in an effort to extend this resolution limit, such as con-focal microscopy, photoactivated localization microscopy, and scanning near fieldmicroscopy. [179]501.6. Fluorescence BackgroundFigure 1.41: Rayleigh criterion for spatial resolution. Top: Airy discs for pointsources. Bottom: Intensity profiles of the discs. (a) Profile of a single point source.(b) Profile of two discs separated at the Rayleigh diffraction limit. (c) Profile of twodiscs where the maxima align with the second minima of the other point, providingclearer resolution. Reprinted from [178].511.6. Fluorescence Background1.6.2 In-situ Fluorescence Imaging of Adsorbed layers at ElectrodeSurfacesFluorescence microscopy is a powerful tool for illustrating the behaviour of sur-factants at the electrode surface. Typical electrochemical measurements such ascapacitance provide some information on the state of the interface, but are limitedto reporting an average across the entire electrode area and are unable to inform onthe fate of any surfactants that are not very near the electrode surface. The use ofin-situ fluorescence provides an opportunity to gain information on not only dif-ferences in behaviour across the electrode area, but to monitor molecules as theymove near the electrode surface, through desorption or other responses to potentialchange.As outlined in Section 1.6.1, if a surfactant layer containing a fluorophore isadsorbed onto an electrode surface, fluorescence will be quenched through res-onance energy transfer into the metal. If the fluorophore is separated from thesurface (as by the ?flipping? of DNA molecules [176, 177] or desorption of thelayer), the fluorophore will be less quenched and fluorescence will return. Thisphenomenon was harnessed in early studies of the electrochemical behaviour of12-(9-anthroyloxy) stearic acid (12-AS). [180] This work showed, by fluorescencemeasurements taken in a similar geometry to the electro-reflectance experiments ofSection 1.5.2, that fluorescence of adsorbed 12-AS (one to three monolayers thick,on Au(111)) was quenched by proximity to the electrode surface. For potentialsat which the layer was desorbed based on electrochemical measurements, fluores-cence signal was observed, both confirming that the layer had been desorbed fromthe electrode surface, and that it remained within the focal volume of the setup, thusdid not move far from the electrode once desorbed. Similar results were found byLi et. al. in studies of monolayers of cystamine tagged with Alexa 488 fluorophore[181, 182].True fluorescence microscopy of an adsorbed layer began with studies of oc-tadecanol mono- and bi-layers. Confocal microscopy studies [115] of octadecanolmono- and bilayers, mixed with a small amount of a fluorescent dye, showed a sim-ilar behaviour to the previously observed 12-AS layers, with a low fluorescence atpotentials where the octadecanol was adsorbed and a sharp increase in fluorescenceas the potential was moved to the range where the octadecanol was desorbed fromthe surface. These studies were extended using traditional fluorescence microscopy521.6. Fluorescence Background[113, 114, 183, 184]. Monolayers of octadecanol, mixed with a small amount ofthe fluorescent dye 1,1?-dioctadecyl-3,3,3?,3?-tetramethylindodicarbocyanine per-chlorate (DilC18(5)), showed fluorescence behaviour correlating with the estab-lished changes in capacitance. [185] Depicted in Figure 1.42a, images of theelectrode surface clearly show a low fluorescence at potentials where the octade-canol layer is adsorbed, and a higher fluorescence at potentials of desorption. Theonset of the increase in fluorescence, more easily visible in Figure 1.42b, corre-sponds to the higher capacitance desorbed state of the layer. The images clearlyshow a non-homogenous surface with many presumed aggregates composing thelayer at desorption. These structures are remarkably stable through several cyclesof adsorption and desorption[184] and are influenced by the structure of the oc-tadecanol layer before adsorption rather than the underlying electrode structure.Casanova-Moreno and Bizzotto [162], Shepherd et al. [185] The fluorescence im-ages of Figure 1.43 show that although there is distortion in the layer structureupon deposition, the overall fluorescence properties of the octadecanol monolayerare largely determined by the structure of the floating layer. This study also pro-vides an example of using the FRET properties of the BODIPY fluorophore toprobe layer structure. The octadecanol layer in these images contains 1 mol % ofthe fluorophore BODIPY-HPC ((2-(4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-dodecanoyl)-1-hexadecanoyl-sn-glycero- 3-phosphocholine), a phos-pholipid modified to include the BODIPY moiety. The octadecanol layers contain-ing this fluorophore were imaged at the wavelength of monomer emission as wellas at the DII dimer emission wavelength. As can be seen in Figure 1.43 (a), thereare monomer-rich and dimer-rich regions of the layer even at formation. Dimer-rich regions are expected to be either enriched in fluorophore, allowing more dimerto form, or more fluid in nature, allowing the monomers to aggregate into dimersmore easily.These same techniques have been applied to other systems of adsorbed surfac-tants. Studies of mixed monolayers of octadecanol and oleyl alcohol show similarbehaviour to the ?pure? octadecanol layers. [183].531.6. Fluorescence Background(a) Differential capacitance (top) and fluores-cence images (bottom). The dashed capacitancetrace (fluorescence images A-H) corresponds tothe initial desorption scan, and the solid trace(images I-P) the return scan. The dotted trace isthe capacitance of the electrode in the absence ofoctadecanol. Reprinted from [185]. CopyrightElsevier 2002.(b) Average grayscale values (a), mean parti-cle area (b) and number of particles (c). Solidsquares: images taken with a sweeping potentialperturbation; open circles: images taken with astepping potential perturbation. Reprinted from[185]. Copyright Elsevier 2002.Figure 1.42: Fluorescence imaging of an octadecanol monolayer with 3 mol %DilC18(5) dye on a Au(111) electrode.541.6. Fluorescence BackgroundFigure 1.43: Fluorescence images of an octadecanol monolayer before adsorption(a) and the same region after adsorption onto a Au(111) electrode and the appli-cation of a potential of (b) -600 mV/SCE and (c) -800 mV/SCE. The monolayercontains 1 mol% of BODIPY-HPC. Scale bars are 50 ?m in all images; top (red)images are of the dimer fluorescence, bottom (green) images of the monomer emis-sion. Reprinted from [162]. Copyright Elsevier 2010.551.7. Atomic Force Microscopy Background1.7 Atomic Force Microscopy Background1.7.1 TheoryAtomic Force Microscopy (AFM) generates images of the sample surface by scan-ning a probe (a pointed tip mounted on the bottom of a flat cantilever) across thesample surface. The cantilever is mounted on the bottom of a piezoelectric tubewhich controls the position of the tip and cantilever in the x, y, and z dimensions tonanometer or better precision. The deflection of the cantilever (up/down as well asleft/right) is monitored via a laser spot reflected off of the back side of the cantileverand onto a photodetector.As the sharp tip is brought near the sample surface, a number of forces act uponit that may affect the imaging qualities. As the tip approaches the surface, it beginsto experience attractive (mostly Van der Walls) forces. At a certain point duringthe approach (point 2 in Figure 1.44), the attractive forces are strong enough toovercome the stiffness of the cantilever, and the tip will ?snap to contact? with thesample surface. Once in contact with the surface, the tip presses into the surface,experiencing a repulsion as the sample compresses. If the tip is once again with-drawn from the surface, it will remain in contact with the surface past the pointat which the initial snap-to-contact occurred. This is partially due to the attractiveforces previously mentioned, but can be enhanced by surface tension of a liquid-coated sample or by spontaneous formation of a water meniscus around the point ofcontact when imaging in humid air which must be broken in order to remove the tipfrom the surface. Because of this, a more rigid cantilever is often used for imagingin air to aid in overcoming the attractive forces on withdrawal. When imaging inliquid, these forces are less severe and a more flexible cantilever may be used.The simplest imaging mode in AFM is ?Contact Mode?, where the probe tipis held in gentle contact with the surface and swept in a raster-scan pattern acrossan area of the surface. The tip is maintained at a constant force on the surfaceby setting a constant cantilever deflection value based on the position of the re-flected laser spot. In order to maintain this position relative to the surface whileaccommodating surface features, the tip is raised and lowered via the piezoelectrictube. This motion provides a height profile as the tip is scanned across the surface,producing a topographical image of the surface. This imaging mode is effective,however since the tip is in constant physical contact with the surface, it is possibleto cause distortion or damage to the sample surface during imaging. This property561.7. Atomic Force Microscopy BackgroundFigure 1.44: Tip-sample interactions at specific points in a force-curve cycle forAFM. Point (1): Large tip-sample separation - no interaction. (2): Tip approach-ing sample and experiencing attractive forces; ?snap-to-contact?. (3) Maximumapproach point; tip presses into sample and experiences repulsive forces. (4) Re-traction; tip no longer indents surface and only attractive forces are felt. (5) Finalpoint of contact, where the tip-sample adhesion force is equal to the restoring forcefrom the bent cantilever. From [186].571.7. Atomic Force Microscopy Backgroundhas been taken advantage of in nanolithography, etching patterns into substrates ata precision difficult to match with other techniques. [187, 188]An alternate method for AFM imaging is intermittent-contact mode, also var-iously called tapping mode, or AC mode. In this technique, the cantilever is os-cillated at or near its flexural resonant frequency, and held just above the samplesurface, such that the cantilever only contacts the surface at the bottom of eachoscillation. The amplitude of the cantilever oscillation is monitored, and the piezo-electric tube is used to maintain the cantilever at a position above the sample wherethe resonant amplitude is constant. Again, in this method, the movement of thez-scanner needed to maintain the constant amplitude provides the topographicalprofile while scanning the surface. This method is typically more gentle to thesubstrate than contact mode, although at high enough oscillation amplitudes it ispossible for the tip to move loosely adhered particles on the sample surface. [189]Typically AC mode imaging uses a piezoelectric oscillator in the cantileverhousing to drive the oscillation of the cantilever and tip. This can create someproblems, especially when imaging in liquid, as the oscillation of the cantileverchip can couple to resonances in the cantilever holder assembly. In liquid, an in-crease in the driving amplitude is required to overcome the solution resistance tocantilever movement, exacerbating the problem. An alternative method for induc-ing the oscillations in the cantilever is ?Magnetic AC? or MAC mode. It is function-ally identical to regular AC mode, however instead of using a piezoelectric actuatorto vibrate the cantilever chip, the cantilever is coated with a thin magnetic coating,and an oscillating magnetic field is generated by a solenoid located directly abovethe cantilever or below the sample. Use of the magnetic field to drive the oscillationavoids many of the issues with noise associated with the typical AC mode, and hasbeen successfully used to image many delicate samples. [190?193]As the oscillation of the cantilever is being monitored during AC mode imag-ing, it is possible to measure its phase shift as well as the amplitude. There is someintrinsic phase lag between the oscillation of the tip and the drive signal being fed tothe cantilever, which can be easily accounted for by typical imaging software. Af-ter accounting for this lag, the phase relation to the oscillation amplitude is shownin Figure 1.45, calculated by modeling the cantilever as a damped driven harmonicoscillator. Near the resonant frequency, the rate of change of the phase with os-cillation frequency is quite large, resulting in enhanced phase contrast with smallamplitude shifts.581.7. Atomic Force Microscopy BackgroundFigure 1.45: Frequency and phase response of a cantilever modeled as a dampedharmonic oscillator. From [186].Damping of the oscillation caused by tip-sample interactions will cause a smallshift in the oscillation frequency. The accompanying phase shift is easily measured,and used for phase-contrast imaging modes. The tip-sample interactions may in-clude (but are not limited to) Van der Walls interactions, electrostatic attractions,or chemical interactions. Changes in tip-sample adhesion as well as sample elas-ticity and viscosity will affect the measured phase lag in the image, probing localvariations in surface properties.Although quantitative analysis of the phase shifts is possible, the presence of anumber of complicating factors makes such analysis difficult. Although a changein the surface interactions will produce a change in phase, a change in phase is notnecessarily due to a change in surface properties. Importantly, the surface topogra-phy will be coupled into the phase contrast. Sharp features (especially indentationswith a radius similar to the tip radius) will have enhanced Van der Walls type attrac-tions between the tip and surface. These attractions are local in nature but do notreflect a change in surface properties, only geometry. This causes a change in thephase along these features, highlighting topographical features. Figure 1.46 showsimages of polycrystalline gold, for which there should be no variance in tip-surface591.7. Atomic Force Microscopy BackgroundFigure 1.46: AFM images of a polycrystalline gold sample. Circled regions high-light narrow regions on the sample where a geometry-induced increase in tip-sample interaction causes a change in the phase offset. From [186].Figure 1.47: AFM height and phase images of a polyvinyl alcohol thin film. Region1 is a strongly adhered layer covering most of the substrate. Region 2 is a highlycrystalline domain approximately 1 nm high, and Region 3 a heterogeneous, thickerlayer approximately 3 nm high. From [186].interaction based on chemical properties. The circled regions highlight valleys inthe surface which create a strong tip-surface interaction that is visible in the phaseimage, despite being chemically identical to the surrounding regions.Similarly, other properties such as surface potential can affect the tip-sample in-teraction, causing a change in phase that is not associated with a region of differentchemical properties. For these reasons, phase imaging is often used qualitatively,to highlight regions with varying tip-sample interactions, rather than for quanti-fying the absolute magnitude of these interactions. Figure 1.47 shows a typicaltopography-phase imaging pair for a thin film of polyvinyl alcohol which has threedistinct domains, differentiable by both the height of the layers and by their phasebehaviour.601.7. Atomic Force Microscopy Background1.7.2 Atomic Force Microscopy of Vesicles and Adsorbed LayersAFM has been used extensively to probe the properties of thin films and surfacelayers. Some of these studies have been discussed in the context of bilayer self-assembly in Section 1.4.1 previously, and will be only briefly mentioned here.Topographical imaging of vesicle-modified substrates has been an invaluabletool in probing the three-dimensional structure of the adsorbed vesicles and self-assembled phospholipid bilayers. As the vesicles and layers are delicate and onlyloosely adhered to the substrate, AC or MAC mode is typically used. Many studieshave used AFM to determine the size of liposomes adsorbed on a solid substrate(reviewed in [194]), although the sizes measured by AFM may be slightly largerthan sizes found in solution due to vesicle deformation when resting on the sur-face. [195] For small unilamellar vesicles, the measured ratio of width to height ofadsorbed vesicles was found to increase to approximately 5, suggesting that somesignificant deformation occurs for vesicles adsorbed intact on glass surfaces. [196]Applications beyond simple size measurements are also possible; the use of ahigh-speed, high force image in contact mode was used to characterize the adhesionof phosphatidylethanolamine bilayers on mica substrates. After modification of themica surface with a single bilayer, this phospholipid will deposit a second bilayeron top of the first. Figure 1.48 shows the AFM image taken of such a surface aftera high-force image of a smaller area was made. Based on the height differences be-tween the area imaged under high-force and the unmodified area, it was shown thateven with repeated imaging, only the upper bilayer would be removed by contactwith the AFM tip.[197]Rigidity of the vesicles adsorbed on a substrate is also measurable by AFM- both by height correlation in traditional imaging methods and by force-distanceanalysis. In the rigidity determination by imaging [198] , the adsorbed liposomesare compared to rigid polystyrene spheres of known radius and their geometry com-pared. More rigid liposomes are assumed to hold a more spherical shape, and werecharacterized by the ratio of their height measured by AFM over their radius asdetermined by dynamic light scattering particle size analysis. Using this method,it was found that the type of phospholipid used to form vesicles and the amount ofcholesterol incorporated into the membrane had an effect on the overall rigidity ofthe adsorbed liposomes.Using force-distance spectroscopy to measure liposome rigidity [199, 200] in-611.7. Atomic Force Microscopy BackgroundFigure 1.48: AFM image (taken in contact mode) of a mica surface modified with apartial-double-bilayer of phosphatidylethanolamine, with the centre of the imagingarea additionally modified by a high-force imaging scan. Comparison of the heightsreveals that the cleared area inside the high-force scan region retains the bottombilayer and only the second bilayer has been moved by the imaging. Reprintedwith permission from [197]. Copyright 1999 American Chemical Society.volves pushing an AFM tip through the vesicle and recording the force required toburst completely through to the underlying substrate. Figure 1.49 shows a sampledata set for the approach and retraction parts of a force-curve cycle, with the two?jump-in? points where the tip pushes through each bilayer clearly visible. Thestiffness of these vesicles can then be calculated from the relationship of force ap-plied to tip position. Measurements performed using this technique showed thataddition of cholesterol to egg phosphatidylcholine vesicles increased the bendingmodulus dramatically.Although in the force curve method, it is possible to easily punch the probethrough an adsorbed vesicle, AFM imaging can be delicate enough to resolve finedetail of adsorbed structures. One example of such an achievement is imagingof so-called lipid rafts in adsorbed mixed-component bilayers. Lipid rafts, or mi-crodomains, are cholesterol and sphingolipid rich regions in a bilayer membrane.These regions are believed to exist in a liquid-ordered phase distinct from theliquid-disordered phase of the surrounding bilayer (Figure 1.50), causing a slightdifference in bilayer thickness between the two regions. Many studies, for exam-ple [201?205], have used atomic force microscopy as a method to investigate theformation and characteristics of these domains. Figure 1.51 shows some AFM to-pography images of a supported sphingomyelin/DOPC mixed monolayer contain-ing various amounts of cholesterol. The slightly taller microdomains are clearlyvisible, although the height difference between the raft and membrane bulk is only0.8 nm or less. Combination of AFM with other techniques, such as fluorescence621.7. Atomic Force Microscopy BackgroundFigure 1.49: Force vs distance curve of an AFM tip placed directly onto an eggphosphatidylcholine liposome. Point 1 corresponds to the initial contact of the tipto the top surface of the liposome. The transition from Point 2 to Point 3 is thebursting of the tip through the topmost bilayer of the liposome into the vesiclecavity. The transition at Point 4 is the bursting of the tip through the bottommostbilayer and into direct contact with the substrate. The interactions on the retractionportion of the curve are more complex. Note, however, that the final jump-out pointis at a distance greater than the liposome radius as it is elongated due to adhesiveforces with the tip during retraction. Reprinted from [199] with permission fromElsevier.631.7. Atomic Force Microscopy BackgroundFigure 1.50: Cartoon schematic of a lipid raft (gray molecules with cholesterol)in a liquid-ordered state surrounded by other membrane phospholipids (greenmolecules) in a liquid-disordered state. Adapted with permission from [202].Copyright 2007 American Chemical Society.microscopy, can provide further information on the bilayer behaviour. In one suchstudy [206] it was demonstrated that a selection of fluorescent dyes will partitioninto different domains that were also identifiable by AFM imaging.1.7.3 In-situ Atomic Force Microscopy with ElectrochemistryAtomic Force Microscopy is easily adapted for imaging under liquid, and in factoften produces superior images due to the lack of the tip adhesion caused by spon-taneous meniscus formation in air. It is then a logical extension to conduct AFMimaging in an electrochemical cell, and in fact a great deal of work has been donecombining potential control with in-situ atomic force microscopy. Of especial in-terest in the context of this work are studies done on solid-supported organic layers.Imaging of the adsorption N-dodecyl-N,N-dimethyl-3-ammonio-1-propanesulfonatefrom a low-concentration solution onto an Au(111) substrate under potential con-trol [207] revealed that at larger negative charge densities (?M < -5 ?C/cm2) a?spongy?-textured monolayer forms, while at higher charge densities(?M > +5 ?C/cm2) a blistered film is favored. At intermediate charge densities,a film containing long stripes of hemimicellar aggregates forms. These structuresare shown in Figure 1.52, along with the structure of the surfactant. Formation ofsimilar long aggregates at positive electrode potentials was previously observed inAFM and STM imaging of bipyridine monolayers on Au(111). [208] In other ex-periments, potential-induced orientation changes of DNA helices on a solid supportwas observed using AFM[209], paving the way for the DNA-based fluorescencesensors under current development. [176, 177]641.7. Atomic Force Microscopy BackgroundFigure 1.51: AFM images of mixed sphingomyelin/DOPC monolayers (1:1mol/mol) containing various amounts of cholesterol. (A) no cholesterol, (B) 10mol% cholesterol, (C) 20 mol% cholesterol, (D) 33% mol cholesterol, (E) 50 mol%cholesterol, and (F) 50 mol% cholesterol monolayer deposited from a subphasecontaining methyl-? -cyclodextrin. All layers are formed on mica by Langmuir-Blodgett deposition. Reprinted from[201] with permission from Elsevier.Figure 1.52: Atomic force microscopy images of N-dodecyl-N,N-dimethyl-3-ammonio-1-propanesulfonate (structure at left) on a Au(111) electrode at large neg-ative charge density (?M < -5 ?C/cm2, leftmost image), intermediate charge density(centre image), and large positive charge density (?M > +5 ?C/cm2, rightmost im-age). At each potential, a characteristic molecular orientation and film structureis observed. Reprinted with permission from [207]. Copyright 2007 AmericanChemical Society.651.7. Atomic Force Microscopy BackgroundFigure 1.53: Atomic force microscopy image (MAC mode, in 0.1 M NaF) of adimyristyl phosphatidylcholine layer under potential control. The fast scan axis isvertical, and the slow scan axis proceeds from left to right across the image. Duringimaging, potential was changed from 0.2 V/AgAgCl to 0.45 V/Ag/AgCl (in thecentre of the image) and back. The potential profile applied is illustrated abovethe image, with the applied potential aligned to the portion of the image recordedat that potential. Below the image is a height profile along the line marked on theimage. Reproduced from [210] with permission from Annual Reviews.661.8. Scope of the ProjectTwo potential-linked states were also found for bilayers of dimyristyl phos-phatidylcholine adsorbed on Au(111). [65, 210] Figure 1.53 shows an AFM imagetaken as the electrode (substrate) potential was scanned from 0.2 to 0.5 V/Ag/AgCland back. The fast scan axis is oriented vertically in this image, so that each col-umn of image data is acquired at a slightly different potential, progressing along theapplied profile as the image progresses from left to right. In the image, it is clearthat the lower potentials have a distinct corrugated layer structure that is altered atthe more positive potentials, adopting a flatter ?melted? structure. The corrugatedstructure reappears as the potential is returned to lower potentials at the right handside of the image.1.8 Scope of the ProjectSolid-supported bilayers of surfactants and phospholipids form a major motif inbiosensor design. Especially for phospholipid bilayers, the ability to incorporateand study transmembrane proteins in an environment where both the chemical andelectrical properties can be controlled is an advantage. Such bilayers are easilyformed by self-assembly from a solution of liposomes with the desired character-istics, however phospholipid bilayers deposited directly on a solid surface havesome notable disadvantages. They do not contain a large hydrated ?cushion? intowhich transmembrane proteins can extend, reducing their usability. As well, theydelaminate from the substrate if removed from the aqueous environment. Somestrategies, such as tethered bilayers[78, 79] and use of spacer polymers [74] havebeen attempted in the past to address these issues.By first modifying the electrode with a layer of insoluble surfactant (Figure1.54a), interaction of liposomes directly with the electrode surface is prevented.Creating pores or defects in this passivating layer (Figure 1.54b) would then allowliposomes to interact with the electrode through these pores (Figure 1.54c), and tobecome incorporated into the adsorbed layer (Figure 1.54d) to produce a final hy-brid bilayer. Depending on the characteristics of the initially adsorbed layer, it maybe possible to produce a hybrid layer where the phospholipid bilayer regions arebuckled or lifted from the surface (as with the hemiliposomal structure illustratedin Figure 1.54d) which would facilitate the use of transmembrane proteins by pro-viding a pocket of electrolyte into which they can extend. The initially adsorbedlayer, if it is itself more stable, may also impart some stability to the DOPC bilayer,671.8. Scope of the Project(a) (b)(c) (d)Figure 1.54: Illustration of the proposed hybrid bilayer formation process. (a) Theelectrode is first covered in a physisorbed layer (here, octadecanol). (b) Defects areintroduced into the adsorbed layer. (c) Liposomes from solution approach the elec-trode and interact with the defect sites. (d) The phospholipid bilayer incorporatesinto the adsorbed bilayer, forming a hybrid layer.preventing it from delaminating on removal from aqueous solution as bilayers ofphospholipids are wont to do.In order for such a system to be feasible, the surfactant layer must be stable onthe electrode surface over a range of conditions, and it must have some mechanismby which to induce defect creation. Octadecanol is well-characterized in electro-chemical systems, and offers all of these advantages. Defect creation is easilyaccomplished by application of a moderate electrical potential (see Section 1.5.2)and is reversible. Using DOPC as a model phospholipid, Figure 1.55 highlights awindow of electrical potential in which defects can be created in the octadecanol,but adsorbed DOPC remains stable on the electrode surface. Within this region(-0.2 V to -0.4 V/SCE) liposomes in solution are predicted to have the best oppor-tunity to incorporate into an existing octadecanol bilayer, as depicted in the cartoonof Figure 1.54.The focus of the work presented in this thesis is the development and char-acterization of a hybrid octadecanol - DOPC bilayer using the scheme describedabove. Electrochemical methods, primarily differential capacitance, along with in-681.8. Scope of the Project05101520253035Capacity (?F/cm2 )DOPC bilayerbare Au(111)05101520253035-1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2Capacity (?F/cm2 )E (V vs. SCE)Octadecanol bilayerbare Au(111)Figure 1.55: Differential capacitance scans during desorption of a DOPC bilayer(top frame) and octadecanol bilayer (bottom). The region highlighted in blue ex-tends from the onset of defect formation in octadecanol (-0.2 V) to the beginningof the first phase change in the DOPC bilayer (-0.4 V/SCE), representing the po-tential range in which liposomes are expected to be most effective at incorporatinginto a solid-supported octadecanol bilayer. Capacitance measurements were per-formed with a 5 mV/s scan rate, 5 mV RMS potential perturbation, and a 25 Hzperturbation frequency.691.8. Scope of the Projectsitu fluorescence and atomic force microscopy are used to observe the system anddetermine its behaviour.70Chapter 2Experimental MethodsThis chapter contains the general procedures and instrumental information for ex-periments presented in the following chapters. In all cases, further experimentaldetail is included alongside the relevant results.2.1 MaterialsAll glassware used was cleaned by heating for two hours in a 1:1 H2SO4:HNO3bath, rinsing in MilliQ (18.2 M?) water, and stored filled with MilliQ water overnight.Electrochemical measurements were performed in a solution of 0.05 M KClO4(Fluka puriss, triply recrystallized) in MilliQ water or in a 0.1 M NaF solution(Sigma SigmaUltra) in MilliQ water. A 3 mg/mL solution of octadecanol (FlukaSelectoPhore) was prepared in chloroform (Fisher, HPLC grade). Octadecanol so-lutions containing various concentrations (0.5 to 8 mol %) of 4,4-difluoro-1,3,5,7-tetramethyl-8-(18-octadecanol)-4-bora-3a,4a-diaza-s-indacene (BODIPY-C19-OH),synthesized in-house[1], were prepared by addition of stock BODIPY-C19-OH so-lution (7 mg/mL in chloroform) to a clean vial, evaporating the solvent under argon,and filling with 3 mg/mL octadecanol solution. The volumes transferred were mon-itored by weighing the vial at each transfer step, and the final concentration of theBODIPY-C19-OH in the solution was verified using absorbance. The BODIPY-C19-OH concentrations reported in the text are the nominal values. Actual concen-trations were within 0.5% of the nominal values (within 0.05% for the 0.5 mol%solution), but exact values are not reported in the text for clarity as several solutionswere used of each concentration throughout the time frame of the experiments.2.1.1 ElectrodesElectrochemical measurements and in-situ fluorescence measurements were per-formed on a polished Au(111) electrode, prepared as described previously [183,712.2. Liposome Formation185]. Before and after each experiment, the electrode was cleaned by flame an-nealing with a butane torch and quenching in MilliQ water, repeating a minimumof three times. A gold coil, used as the counter electrode, was prepared using thesame procedure.2.2 Liposome FormationLiposomes were formed using the extrusion method [211], using a commercial ap-paratus (LIPEX Extruder, Northern Lipids Inc.) fitted with a polycarbonate filter(Nucleopore 0.1 ?m, Costar Corporation). Dioleoyl phosphatidylcholine (DOPC,in chloroform, Avanti Polar Lipids, Inc.) was dried under argon (Supelco) from theas-provided chloroform solution, then sonicated to form a suspension in 0.1 M NaFelectrolyte . This suspension was cycled through the extruder under high-pressureargon ten times to form a monodisperse suspension of liposomes. The resultingsolution was stored in a closed glass vial under refrigeration. After formation, thesize distribution of liposomes was measured with dynamic light scattering (Coul-ter N4+ particle size analyzer). This procedure and results of particle sizing aredesccribed in Appendix B.2.2.1 Chemical Stability of LiposomesAfter formation and periodically throughout the lifetime of the solution, the solu-tion was tested for lipid degradation by thin-layer chromatography (TLC). The li-posome solution was spotted (1 ?L) onto an aluminum-backed silica gel TLC plate(EMD) along with a sample of stock DOPC solution in chloroform (0.5 ?L) anda known degraded liposome solution (1 ?L) in adjacent lanes. The degraded lipo-some solution was prepared from liposomes extruded in the perchorate electrolyte,saturated with oxygen after preparation, and aged for several weeks. After apply-ing the solution to the TLC plate, the spots were allowed to dry thoroughly andthen eluted using a 2:1 chloroform:methanol mixture (both: Fisher, HPLC grade).After elution, the plate was dried and visualized by rinsing in dilute sulphuric acid(Fisher ACS, approximately 1 M) and charring in an oven at 190 C. Multiple spotswas characteristic of degraded phospholipid material, thus only liposome solutionsproducing a single TLC spot were used for electrochemical analysis. Details of themonitoring by TLC are provided in Appendix A.722.3. Surface Pressure Measurements2.3 Surface Pressure MeasurementsMeasurements of the floating monolayers of octadecanol were made during com-pression and at equilibrium spreading pressure (ESP), both independently and withsimultaneous fluorescence imaging.2.3.1 Compression Isotherms and ESPCompression isotherms of the octadecanol monolayers were measured on a Lang-muir trough (Nima model 602A) in the lab of Dr. Guillaume Bussiere. The troughwas prepared by wiping clean with isopropanol and a laboratory tissue, then rins-ing with deionized water. The trough was then filled with the 0.05 M KClO4 elec-trolyte and tested to ensure that the bare water pressure value was accurate and thatcompression of the barrier did not change the surface pressure in the absence of afloating monolayer. Once the trough was prepared and confirmed clean, approx-imately 20 ?L of octadecanol solution was deposited onto the electrolyte surfaceand the solvent allowed to evaporate. Pressure was measured while compressingthe barrier at 150 cm2/min, slowing to 50 cm2/min and 20 cm2/min as surface areadecreased.Measurements of the ESP of the various concentrations of BODIPY-C19-OHin octadecanol were taken by partially filling a 100 mL beaker with the perchlo-rate electrolyte and assembling a microbalance (Nima, model P54) with Wilhelmyplate slightly off-centre of the surface to avoid the floating octadecanol crystal..Octadecanol solution was added in 1-5 ?L aliquots to the surface and the pressureallowed to equilibrate after the chloroform solvent evaporated. Aliquots of octade-canol were added until the measured surface pressure reached a maximum and avisible crystal of excess octadecanol was present at the interface. In some cases,the layer was then ?re-formed? by depositing approximately 50 ?L of chloroformonto the electrolyte-air interface with a microliter syringe and allowing it to evap-orate, consolidating any small crystallites on the surface into one large structure.Surface pressure was considered an ?equilibrium? surface pressure if it remainedstable, within 0.2 mN/m, for at least 200 seconds.732.3. Surface Pressure Measurements2.3.2 In-situ Fluorescence MeasurementsA modified Langmuir trough with an optical window below the subphase (Nimamodel 102M) was used with an epi-fluorescence microscope to obtain fluorescenceimages of the floating monolayers of octadecanol/BODIPY-C19-OH mixtures (Fig-ure 2.1a). The trough was prepared in a similar manner to that described above,with special care to keep the optical window clean and streak-free. To accommo-date the time required for image acquisition, rather than a continuous compression,the layer was compressed by several mN/m then allowed to rest while images weretaken. To collect the fluorescence images, a Photometrics Evolve 512 CCD camerawas used (16 ?m x 16 ?m pixel size), Peltier cooled to -80 C. Exposure time andelectron-multiplier gain settings were varied from experiment to experiment; thesesettings are noted with the images. Illumination was provided by a broad-spectrumsource (Exfo X-cite Exacte), passed through a filter cube custom-assembled to suiteither the monomer or dimer fluorescence of BODIPY-C19-OH (transmission pro-files shown in Figure 2.2).Monomer and dimer images were taken sequentially; monomer first, followedby dimer, using a 10x objective (Olympus LMPlanFl, NA=0.25). The lamp shutterwas closed between images to minimize photobleaching. As the trough was im-mobile, only a small area was available for imaging. To obtain a sampling of thestructures presesnt, monomer-dimer image pairs were taken several minutes apartto allow different areas of the floating layer to drift into view when possible.Images of the floating layer at ESP were taken using the electrochemical cell onthe same microscope assembly (Figure 2.1b). The cell was filled with electrolyteto a depth of approximately 5 mm, and the electrolyte was purged of oxygen bybubbling with argon. During measurements, an argon atmosphere was maintainedin the cell by a minimal flow of argon gas so as to avoid disturbing the layer dur-ing imaging. A 20 ?L aliquot of octadecanol/BODIPY-C19-OH solution was de-posited onto the electrolyte surface with a microliter syringe and the solvent al-lowed to evaporate. Several images of the layer were then taken, following thesame procedure as in the Langmuir trough, but as the cell was mounted on a mov-able stage, it was possible to take images of a sampling of the different structuresacross the electrolyte surface.742.4. Preparation of Modified Electrodes(a) Langmuir Trough on microscope (b) Cell on microscopeFigure 2.1: Epi-fluorescence microscope configuations used for in-situ fluores-cence imaging.2.4 Preparation of Modified Electrodes2.4.1 Liposome-coated ElectrodesTo modify an electrode with a phospholipid bilayer, an electrochemical cell wasprepared with 0.1 M sodium fluoride electrolyte. The freshly flame-annealed andcooled Au(111) electrode was introduced into the cell in a hanging meniscus con-figuration and held under potential control. Liposomes were introduced directlyinto the electrolyte through the cell?s injection port (Figure 2.3) using a microlitersyringe, to a concentration of 30 ?g DOPC/mL electrolyte. The liposomes wereallowed to diffuse through solution and assemble on the electrode surface until aminimum capacitance value was reached (typically one hour), while differentialcapacitance was monitored (see Section 2.5.2 for settings). To simulate the condi-tions that would be required in the fluorescence-type cell, the electrolyte was notstirred.In some cases, the electrode was modified a second time from the same liposome-electrolyte preparation by removing the electrode, flame-annealing, and re-introducingit to the electrochemical cell already containing liposome solution. In these cases,substantially less time was required to reach the minimum capacitance value (ap-proximately five minutes), though measurements continued for the same hour-long752.4. Preparation of Modified Electrodes020406080100400 450 500 550 600 650 700% TransmissionIntensityWavelength (nm)Excitation: Chroma ET470/40xEmission: Chroma ET525/50MDichroic: Chroma T495LPXRBODIPY: AbsorptionBODIPY: Emission(a) Transmission characteristics of filters used for BODIPY-C19-OH monomer fluores-cence.020406080100400 450 500 550 600 650 700% Transmission Wavelength (nm)Excitation: Semrock FF01-482/35-25Emission: Semrock FF01-641/75-25Dichroic: Chroma T585LP(b) Transmission characteristics of filters used for BODIPY-C19-OH dimer fluorescence.Figure 2.2: Transmission characteristics of filter cubes used for fluorescence mea-surements.762.4. Preparation of Modified Electrodes(a) Standard electrochemical cell (b) Electrochemical cell for in-situ fluorescenceFigure 2.3: Schematics of electrochemical cells used, shown with Au(111) elec-trode in hanging meniscus configuration. WE=working electrode, CE=counterelectrode, RE=reference electrode.timescale.2.4.2 Octadecanol-coated Electrodes: Single-touch (monolayer)Adsorbed monolayers of octadecanol were prepared on the Au(111) electrode byLangmuir-Schaefer deposition. A monolayer of octadecanol or octadecanol/BODIPY-C19-OH mixture at ESP was prepared in the electrochemical cell by depositingseveral (20-30) ?L of the relevant solution onto the electrolyte surface and allow-ing the solvent to evaporate under a flow of argon, leaving a small crystal of excessoctadecanol on the surface. The freshly flame-annealed and dried electrode wasintroduced into the cell and allowed to cool in the argon atmosphere for approx-imately two minutes, then touched to the electrolyte surface and lifted to form ahanging meniscus while under potential control. For electrochemical studies ofoctadecanol/BODIPY-C19-OH mixtures, the 0.05 M KClO4 electrolyte was typ-ically used. Where relevant, the type of electrolyte used is noted in the Resultssections.772.4. Preparation of Modified Electrodes2.4.3 Octadecanol-coated Electrodes: Double-touch (bilayer)Electrodes with a bilayer of octadecanol were prepared using sequential Langmuir-Schaefer depositions. First, a floating monolayer of octadecanol or octadecanol/BODIPY-C19-OH mixture at ESP was prepared in a 50 mL beaker or small crystal-lization dish containing electrolyte. An excess of octadecanol solution (20-50 ?L)was deposited onto the electrolyte surface, and the chloroform allowed to evaporate.The beaker or dish was kept covered with a loose lid when not in use. A secondfloating monolayer of octadecanol was prepared in the electrochemical cell, as de-scribed in Section 2.4.2. The Au(111) electrode was flame-annealed and cooled,then touched to the first octadecanol layer (prepared in the beaker) and slowly with-drawn. Any remaining droplet of electrolyte was carefully wicked away with theedge of a laboratory tissue. The electrode was then placed into the electrochemicalcell and allowed to sit in the argon atmosphere for 30 sec, then touched to the elec-trolyte interface and pulled into a hanging meniscus configuration. This methodproduces an electrode which is expected to have a bilayer of octadecanol similar tothe ?X type? of bilayer described in the literature.[46]Composition of bilayer leafletsFor investigations of BODIPY-C19-OH in Chapter 3, the composition of both theex-situ (in beaker) and in-situ layers (in the electrochemical cell) were varied frompure octadecanol to a maximum concentration of 8 mol% BODIPY-C19-OH inoctadecanol. This created both symmetrical bilayers, where both leaflets had thesame composition, and asymmetrical bilayers, where one leaflet was pure octade-canol and the other contained the BODIPY-C19-OH fluorophore.For the investigations of liposome interaction with octadecanol bilayers de-tailed in Chapters 4 and 5, a single composition was used. In these studies, theex-situ layer was composed of a 3 mol% mixture of BODIPY-C19-OH in oc-tadecanol, and the in-situ layer of pure octadecanol. The fluorophore-containinglayer was thus deposited directly onto the electrode surface, minimizing the metal-fluorophore separation and ensuring that the fluorescence was maximally quenchedwhild adsorbed.782.5. Electrochemical Methods2.4.4 Multiple Depositions From Octadecanol MonolayersDue to the solid nature of octadecanol at room temperature, the number of good-quality Langmuir-Schaefer depositions that can be performed from a single floatingmonolayer is limited by the relative surface area of the substrate and of the floatingmonolayer. For example, it was found during these investigations that a maximumof two such depositions onto the Au(111) electrode could be reliably made froma monolayer prepared in a 50 mL beaker. After this point, the monolayer mustbe re-formed or a new monolayer deposited in order to maintain a consistent layerquality.For monolayers composed of pure octadecanol, the monolayer can be re-formedby careful application of chloroform to the electrolyte surface, dissolving the oc-tadecanol at the interface and allowing it to redistribute into a monolayer with thesame properties as a freshly deposited one. This reformation of the layer is not suit-able for monolayers containing the BODIPY-C19-OH fluorophore, however. Forreasons that were not investigated here, the addition of chloroform causes an unevendistribution of fluorophore in the re-formed layer and large non-fluorescent regionsare formed. For this reason, for every two ex-situ depositions of fluorophore-dopedoctadecanol required for an experiment, a fresh layer was formed in a clean beaker.For experiments where the BODIPY-C19-OH fluorophore was located in the elec-trochemical cell, it was not possible to re-form the layers, so the experiment lengthwas limited to one or two depositions, depending on the size of cell used. For theinvestigations of liposome interaction with the octadecanol bilayers, pure octade-canol was used in the electrochemical cell in order to facilitate multiple depositionsin one experiment day without conflict caused by fluorophore leaching after layerreformation.2.5 Electrochemical MethodsElectrochemical measurements were performed in electrolyte purged of oxygen bybubbling with argon, and maintained under positive argon pressure during the ex-periment. A gold wire coil was used for the counter electrode, and a saturatedcalomel electrode (SCE) connected via a salt bridge was used as a reference. Thesalt bridge was filled with electrolyte, allowing electrical connectivity while thestopcock prevented mixing of solutions and chloride contamination of the elec-792.5. Electrochemical Methodstrolyte by the SCE. Figure 2.3 shows a schematic of the typical cell set-up, in-cluding the hanging meniscus configuration used to ensure only the polished (111)face of the electrode was used for electrochemical measurements. Electrochemicalmeasurements were taken with a potentiostat and lock-in amplifier (electrochemicalstation: HEKA PG590 with PAR 175 scan generator and EG&G 5210 lock-in. Flu-orescence station: FHI-ELAB 0599 with EG&G 5208 lock-in) and recorded usingcustom National Instruments LabView software. Experiments outlined in Chapter3 concerning the effect of BODIPY-C19-OH on octadecanol layers were performedin the 0.05 M potassium perchlorate electrolyte. For the investigations of liposomeinteraction with octadecanol (Chapters 4, 5, and6), experiments were performed inthe 0.1 M sodium fluoride electrolyte to prevent oxidation of the phosopholipids.After assembling the electrochemical cell and purging it of oxygen, cyclicvoltammogram (CV) and differential capacitance measurements were taken of the?full-scale? and ?double-layer? potential windows to ensure the cleanliness of theelectrolyte (described below).2.5.1 Cyclic VoltammetryCyclic voltammetry was performed at a scan rate of 20 mV/s. For unmodified elec-trolytes in potassium perchlorate electrolyte, ?full scale? CV limits were +1.25 V to-0.8 V, bounded at the positive limit by bulk gold oxidization and by hydrogen gen-eration at the negative limit. ?Double layer? CV limits, used for characterizing theelectrolyte cleanliness and for comparison with modified electrode, were +0.65 Vto -0.8 V. In the sodium fluoride electrolyte, the full scale and double layer limitswere +1.1 V to -0.8 V and +0.55 V to -0.8 V, respectively, due to the different pHin this electrolyte.After modification, electrodes were first scanned in a range where the adsor-bates are stable and free from defect formation or oxidization: +0.15 V to -0.15 V.After modification with DOPC or octadecanol layers, the electrode was scannedin this range until the current was stable and any wetting on the sides of the elec-trode from the hanging meniscus was minimized. After the CV was stable, furthercharacterization by differential capacitance or CV measurements with an expandedpotential range could be performed.802.5. Electrochemical Methods2.5.2 Differential CapacitanceDifferential capacitance measurements were performed with a perturbation of 5 mVRMS and a scan rate of 5 mV/s where applicable. For standard electrochemicalmeasurements, a frequency of 25 Hz was used. For in-situ fluorescence measure-ments where a stepping potential profile was used, the frequency was increased to200 Hz in order to allow the lock-in amplifier to acquire a stable reading in a shortertime after each step. At the potentials under investigation, this change in frequencyhas a minimal effect on the capacitance measurements.All modified electrodes were characterized by measuring the differential ca-pacitance in the +0.15 to -0.15 V potential range. For investigations of liposomeincorporation into octadecanol layers (described in Chapters 4 and 5), in order toensure all measurements were made on comparable layers, any layers with a min-imum capacitance greater than 1.08 ?F/cm2 were discarded as defective deposi-tions. For other experiments, all layers were used as-formed. Once the capacitancein the +0.15 to -0.15 V range was known, further modification of the layer andcapacitance measurements at extended potential ranges could be performed.2.5.3 Incorporation of LiposomesAfter modification of the electrode with a bilayer of octadecanol, liposomes wereintroduced directly to the electrolyte using a microliter syringe through the injec-tion port of the electrochemical cell (see Figure 2.3). In the larger electrochemicalcell, the liposome concentration was approximately 30 ?g DOPC/mL electrolyte,while in the smaller cell used for in-situ fluorescence measurements, the concen-tration was higher, approximately 45 ?g DOPC/mL electrolyte. As when formingliposome layers directly on the electrode surface, the electrode was held for 60 minat 0 V/SCE while the liposomes diffuse through the electrolyte and differential ca-pacitance measurements were made.After the hour, a series of potential steps was applied to the electrode whilecontinuous capacitance measurements were made. The magnitude and duration ofthe steps was varied, but generally involved a maximum of four stages, as depictedin Figure 2.4.After the application of the potential step profile, the liposome-modified layerwas characterized by a scan to desorption of the octadecanol layer (from +0.15 to-0.8 V/SCE) while measuring capacitance.812.5. Electrochemical Methods-0.8-0.6-0.4-0.200 15 30 45 60E (V/SCE)Time(a)-0.8-0.6-0.4-0.200 15 30 45 60E (V/SCE)Time(b)-0.8-0.6-0.4-0.20E (V/SCE)Time(c)Figure 2.4: Potential step profiles used in testing liposome incorporation. In exper-iments following 2.4c, the time at each potential varied from 1 to 45 min, and oneof the three ?poration potentials? shown (-0.4, -0.6, and -0.8 V/SCE) was chosen.2.5.4 In-situ Fluorescence MethodsIn order to further characterize the potential dependent behaviour of octadecanoland liposome-modified octadecanol layers at the electrode surface, in-situ fluores-cence images were also taken. The electrochemical procedures were the same asdescribed in Section 2.5 with a few minor modifications. To allow image acquisi-tion at a fixed potential, scanning experiments were performed using a ?staircase?profile rather than a smooth potential ramp as in the simple electrochemcial exper-iments. To simulate the 5 mV/sec sweep rate used in the electrochemical experi-ments, a potential step of 25 mV was used with a step width of 5 seconds. At thebeginning of each step, the acquisition of a fluorescence image and a capacitancemeasurement was triggered. Each capacitance measurement was the average of100 individual samples taken in succession. The potential applied was set by theLabView software, and image and capacitance data acquisition was triggered bythe same software at the beginning of each step.While holding at a fixed potential, such as during the 60 min diffusion timeor during the application of the potential step profile for liposome incorporation,image and capacitance measurement were triggered at a fixed interval (5 sec, 1 min,or 5 min, as indicated with each dataset).The camera and light source used was the same as described in Section 2.3.2,however a 50x objective was used for all measurements (Olympus, LMPlanFL,NA=0.5). For the octadecanol/BODIPY-C19-OH studies, a 2.5 sec exposure timewas used with an electron-multiplier gain of 200 and an 80% lamp intensity. Forthe studies on liposome incorporation, two imaging settings were used: 2.5 sec ex-posure and a gain of 200, or a 5 sec exposure time with a gain of 400, both with822.6. Fluorescence Image Analysisan 80% lamp intensity. The exposure times are specified in the data presentations.The longer exposure and higher gain was used to increase signal from the mostlyquenched, adsorbed layer. The lamp shutter was closed in between image acquisi-tions in order to minimize photobleaching.2.6 Fluorescence Image AnalysisAs the specific analysis performed varies considerably across the experiments dis-cussed in this work, a summary of general techniques used is given here, and detailsof the analysis are included in the respective results sections. Image processingwas performed using the ImageJ [212] software suite. All images were initiallyprocessed by applying the built-in Despeckle routine to remove single pixel noise,followed by a 2-pixel radius Median or Accurate Gaussian Blur filter and a subtrac-tion of the dark signal (500 counts). The average grayscale value of the image wastaken and this value is reported on the ?intensity? plots.Image sequences were aligned using the StackReg plugin in ImageJ to accountfor a slight drift in electrode position with time. As this plugin subtracts the mini-mum intensity value from the first image, this quantity was measured and re-addedto the image after alignment. Where a fluorescent background was to be subtracted,the ?Subtract Background? routine, which implements a rolling ball filter [213] ofa given radius, in this case, 50 pixels. When required, in order to compare imageintensities taken at different exposure times in the liposome incorporation experi-ments, images taken at the higher gain setting (400) were divided by an empiricallydetermined conversion factor to convert the values to the equivalent of the lowestgain value used (200). The average intensity value for each image is thus reportedin ?kilocounts per second? (kcts/sec). The derivation of the calibrations factorsused is shown in Appendix B.2.7 Atomic Force Microscopy Methods2.7.1 InstrumentationAn Agilent 5500 atomic force microscope (AFM) equipped with a large scanner(84 ?m ? 84 ?m maximum imaging area) was used to obtain all measurements,using the manufacturer?s provided PicoView software suite. The AFM is housed in832.7. Atomic Force Microscopy Methodsan acoustic isolation box, supported on a lead plinth suspended from bungee cordsfrom the box ceiling in order to minimize vibrational noise from the laboratorysurroundings. The instrument is equipped with a video camera, which was used toaid in alignment of the cantilever tip with the desired region of the substrate and forrecording a reference of the imaging area location.In-situ AFM measurements were done in the electrochemical liquid cell (Fig-ure 2.5b). An environmental chamber was used with a positive-pressure flow ofnitrogen gas in order to provide a clean and oxygen-free atmosphere. As the en-vironmental chamber is not airtight, and the nitrogen gas is not bubbled directlyinto solution in the liquid cell, oxygen may not be completely removed from theelectrolyte during imaging. However, this is not expected to affect the experimentspresented here, as oxygen is not significantly electrochemically active at the poten-tials used.A ?-Autolab potentiostat (Metrohm) was used to provide potential control forin-situ measurements, controlled via the manufacturer?s provided software. A goldbead pseudo-reference electrode was used, with an Au coil counter electrode. Al-though the Au bead is not a true reference electrode, it provided a stable enoughreference to frunction in the electroyte and potential ranges used. This reference,based on the position of the oxide peaks of a bare Au CV in the same electrolyte,was approximately -0.1 V/SCE.2.7.2 SubstratesAtomic force microscopy was performed on either a gold bead electrode or a gold-on-mica substrate. Non-electrochemical measurements were also performed usingthe back of a gold-coated mica substrate, where the imaging surface was freshlycleaved mica, prepared by lifting off the top layer of mica using a fresh piece ofScotch tape (3M) and dusting away stray mica flakes with a stream of dry argon.The gold bead electrode was prepared by heating a gold wire with a butanetorch until it melted to form a bead, then allowing the bead to cool slowly in theflame, forming large flat facets on the bead surface. The bead was then electropol-ished in 0.1 M perchloric acid (Fisher Ultima) by application of an 100 mA/cm2current for 30 seconds, followed by a rinse in 0.1 M hydrochloric acid (FisherACS) and a rinse in MilliQ water. This polishing procedure was repeated threetimes. Before and after each experiment, the bead was cleaned by flame annealing842.7. Atomic Force Microscopy Methods(a) (b)Figure 2.5: Assembly of AFM electrodes used for imaging. 2.5a shows the Aubead with wire inserted into the Teflon block and mounted into sample plate.2.5b is a top-view of the AFM electrochemical cell used for in-situ measurements.WE=working electrode, CE=counter electrode, RE=reference electrode. The WEis Au on mica, and forms the base of the cell, held in place by clamps fixed to thecell plate. The CE and RE are gold wires.with a butane torch, maintaining a red heat for at least five seconds, then quenchingin MilliQ water. The bead was annealed at least three times in this manner, thengently dried in the butane flame and allowed to cool in air before use.Once formed, the bead was cleaned before use by the flame annealing proce-dure described in Section 2.1.1. After any necessary surface modifications wereperformed (see below), the bead was aligned to present a horizontal Au(111) facetfor imaging.The electrode stem was inserted into a Teflon sample holder (Figure2.5a) and a laser spot was reflected from the facet. The bead alignment was man-ually adjusted to within 1? of horizontal by bringing the reflected laser spot intoalignment with the laser source. The substrate was then mounted in the AFM andready for imaging.For in-situ AFM measurements, a gold on mica substrate (Agilent Technolo-gies) was prepared by flame annealing with a butane torch. The substrate wasplaced on a clean quartz plate and the flame passed over it several times over aperiod of approximately 15 seconds. The substrate was allowed to cool for severalseconds in air on the plate before use. It was then mounted in the liquid cell (Figure2.5b), which was cleaned using the same procedure as for the glassware, using aH2SO4?HNO3 bath and storage in MilliQ water, with the exception of the O-ring,which was cleaned by rinsing in methanol and MilliQ water.852.7. Atomic Force Microscopy Methods2.7.3 Surface ModificationFor AFM analysis, octadecanol monolayers were deposited onto Au bead elec-trodes or Au/mica substrates using a method similar to that described in Section2.4.2. The substrates were cleaned and prepared as described above. A floatingmonolayer of octadecanol or a 3 mol% mixture of BODIPY-C19-OH in octade-canol at ESP was formed on a small crystallization dish of electrolyte (either 0.1 MNaF or 0.05 M KClO4) in a method similar to that described for the electrochem-ical cell above, excepting that the dish was not in an argon atmosphere but in air,protected with a loose-fitting lid. The electrode was then oriented so that the flatsurface was parallel to the electrolyte surface, gently touched to the electrolyte, andslowly removed. Any small droplet of electrolyte remaining on the electrode waswicked away with the corner of a laboratory tissue (Kimwipe Delicate Task Wiper,Kimberley-Clark), if necessary aided by first moving it to the edge of the electrodewith a gentle flow of argon.Bilayers for AFM imaging (Au bead and Au/mica substrates) were prepared ina similar manner. The floating monolayers were prepared in a larger crystallizationdish (diameter approximately 10 cm) to ensure enough monolayer surface area wasavailable to perform two sequential depositions. The flat imaging area of the sub-strate was touched twice to the surface using the method described above. Betweentouches, the droplet of electrolyte was wicked away, and the second touch was per-formed on a region of the surface geometrically as far as possible from the area ofthe first touch in order to obtain the best quality deposition possible. After the sec-ond deposition, the droplet of electrolyte was again wicked away from the surface,and the substrate dried under a stream of Ar for several seconds before assemblyinto the AFM. When imaging in air, the electrode was also allowed to dry furtherby resting, mounted in the AFM in the isolation box, for approximately 30 min.2.7.4 Imaging ConditionsAll images of modified electrode surfaces were taken using intermittent-contactmode imaging (AC or MAC). Images for fluorescence correlation and basic layercharacterization were taken in air, using either the Au bead electrode or the Au/micaelectrode. All in-air images were taken in AC mode, using silicon probes with anominal resonant frequency of 190 kHz (VistaProbes, model T190R), tuned to a1.5 V amplitude and -0.1 kHz off-peak, using 80% of the free oscillation amplitude862.7. Atomic Force Microscopy Methodsas the setpoint. Images in solution were taken in MAC mode, using Type II MACcantilevers (Agilent, nominal force constant 2.8 N/m, resonant frequency 75 kHz inair) or Type I MAC cantilevers (Agilent, nominal force constant 0.6 N/m, resonantfrequency 75 kHz in air). The same tuning parameters were used with these tips aswith the VistaProbes tips. For most images, a scan frequency of 1 Hz was used anda resolution of 256 or 512 lines/image. For the large 84 ?m images, a scan rate of0.1 Hz was used and a resolution of 2048 lines/image.87Chapter 3Effects of Fluorescent Dye onOctadecanol MonolayersThe multicomponent bilayer model under investigation here (see Section 1.8) re-lies on using an adsorbed layer of octadecanol covering the electrode surface whichboth prevents liposomes from interacting with the electrode surface in an uncon-trolled manner, and provides the means to create defect areas into which they willlater incorporate. In order for potential controlled incorporation of liposomes toproceed in this model, it is important to characterize the initially adsorbed octade-canol layer. This will be accomplished with electrochemical measurements and viain-situ fluorescence microscopy. As neither octadecanol nor the DOPC used in theliposomes is natively fluorescent, a small percentage of a fluorescent dye moleculemust also be incorporated into the adsorbed layer.Adding a second component to the monolayer runs the risk of disrupting thelayer structure, creating defects and making the depositions less reproducible. Eventhough the BODIPY-C19-OH fluorophore was chosen for its structural similarityto octadecanol, its presence will interfere with the organization of the octadecanolmonolayer to some degree. The concentration should be kept at a minimum in or-der to maintain a good quality adsorbed monolayer. However, the need to obtainthe sufficient signal from the fluorescence imaging demands that the fluorophoreconcentration in the layer be kept as high as possible. Especially for studies of theadsorbed layers, where most of the fluorescence will be quenched by energy trans-fer into the nearby metal, maintaining a strong fluorescence signal is important.These two conflicting demands require that an optimum fluorophore concen-tration is used, one that provides sufficient fluorophore in the monolayer while re-taining the structural and electrochemical properties of the pure octadecanol mono-layer. The studies presented in this chapter will characterize the effects of BODIPY-C19-OH on an octadecanol monolayer, and direct the selection of the fluorophoreconcentration that is most suitable for use in the liposome incorporation studies.883.1. Compression Isotherms3.1 Compression IsothermsOne of the simplest methods to characterize the properties of thin monomolecu-lar films is by measurement of the compression isotherms relating film pressureto the mean molecular area at the surface. The measurement of these isothermsfor octadecanol monolayers with varying concentrations of BODIPY-C19-OH willmeasure the influence of the fluorophore on the layer structure.3.1.1 Experimental MethodsA large surface area Langmuir trough was cleaned and prepared as described inSection 2.3.1 using as the subphase 0.05 M KClO4 electrolyte. Several ?L of anoctadecanol solution containing 0 to 8 mol% of BODIPY-C19-OH were depositedonto the air-solution interface and the solvent allowed to evaporate. The barrierwas closed at 150 cm2/min, slowing to 20 cm2/min as surface area decreased, whilemeasuring the film pressure.The equilibrium spreading pressures (ESP) of the monolayers were measuredseparately by spreading a small amount of the octadecanol/BODIPY-C19-OH solu-tion onto the air-solution interface of a beaker filled with 0.1 M KClO4 electrolyteand monitoring film pressure as more of the octadecanol mixture was added, untila maximum film pressure was reached. Details of this procedure are also found inSection ResultsTypical compression isotherms for each fluorophore concentration are presented inFigure 3.1. The isotherm for pure octadecanol (black line) closely resembles liter-ature reports for the octadecanol isotherm (Figure 1.4, [35]). As the concentrationof BODIPY-C19-OH increases, a bump appears at approximately 15 mN/m, whichdevelops into a plateau at higher BODIPY-C19-OH concentrations. The appear-ance of a plateau at intermediate film pressures is typical of layers containing twocompletely or partially immiscible components, and suggests that the BODIPY-C19-OH is not distributed evenly in the octadecanol monolayer,Measurements of the ESP of the octadecanol/BODIPY-C19-OH layers (Figure3.2) shows a minimum in ESP at approximately 3 mol%. This may be due tosome change in layer structure between the low and high concentrations. In the893.2. Electrochemical CharacterizationFigure 3.1: Compression isotherms for monolayers of octadecanol containing 0, 1,3, 5, and 8 mol% BODIPY-C19-OH. Film pressure range plotted is limited to theregion below the minimum collapse pressure for all layers.isotherms, the plateau at approx. 15 mN/m is much more pronounced in the 5and 8 mol% than might be expected from the trend in the 1 and 3 mol% layers,supporting the observation that there is a change in behaviour around the 3 mol%concentration. This observation is supported by electrochemical study of theselayers, discussed next.3.2 Electrochemical CharacterizationExamination of the electrochemical behaviour of the octadecanol/BODIPY-C19-OH layers will be influenced by the character of the monolayers formed at theair-solution interface. A layer with many defects will, when deposited onto theelectrode, show a higher minimum capacitance value, and different desorption char-acteristics in the cyclic voltammogram and differential capacitance scans. As well,it is important to ensure that the layer chosen to use for the liposome interactionstudies can be formed reproducibly from the floating monolayer and has consistentdesorption and poration characteristics.903.2. Electrochemical CharacterizationFigure 3.2: Equilibrium spreading pressure (ESP) and minimum capacitance valuesfor octadecanol monolayers containing 0, 0.5, 1, 3, 5, and 8 mol % BODIPY-C19-OH fluorophore.3.2.1 Experimental MethodsElectrode preparation and electrochemical measurements were performed accord-ing to the general procedures outlined in Sections 2.4.1 and 2.5. For these studies,various solutions of octadecanol containing different amounts of the BODIPY-C19-OH fluorophore were prepared for use in forming the floating monolayers usedfor deposition from the floating monolayer onto the electrode. Monolayers of oc-tadecanol were deposited onto the electrode using the Langmuir-Schaefer methoddescribed in Section 2.4.1. Bilayers were formed by performing two sequentialLangmuir-Schaefer depositions from the floating monolayer in the electrochemicalcell. Between the first and second depositions, the electrode was removed from thecell, and the droplet of electrolyte remaining on the electrode surface gently wickedaway with the edge of a laboratory tissue. The electrode was then re-introduced intothe electrochemical cell, allowed to equilibrate in the Ar atmosphere for approxi-mately 5 seconds, and then touched to the electrolyte surface to form the secondLangmuir-Schaefer deposition.All measurements are performed in the hanging meniscus arrangement, en-suring that the electrode area is constant across measurements, and that only theAu(111) face is exposed to solution. After deposition, each octadecanol mono-913.2. Electrochemical Characterizationor bi-layer was monitored by cyclic voltammetry between +0.15 V/SCE and -0.15 V/SCE until subsequent scans overlapped, an indication that the edges of theelectrode had dried . The potential scan limits were then increased to +0.15 V/SCEto -0.8 V/SCE and the layer cycled until an equilibrium response, where sequentialscans overlapped exactly. This equilibrium CV was recorded, then a differentialcapacitance measurement was made using the same potential range. As mentionedin Section 1.5.2, significant changes in the layer occur, especially between the firstand second potential sweeps. Recording the equilibrium scans removes any uncer-tainty in comparing the sequentially acquired CV and capacitance scans.3.2.2 ResultsThe CV and capacitance scans for monolayers of the BODIPY-C19-OH concen-trations tested are shown in Figure 3.3. The leftmost (black) scans show an exam-ple of pure octadecanol, with no fluorophore added. As the concentration of theBODIPY-C19-OH increases, deviations from this ?ideal? behaviour become morepronounced. At 0.5 mol%, a small shoulder appears on the tallest pseudocapaci-tance peak of the adsorption scan (at approx. -0.2 V/SCE), showing that there issome disruption in the octadecanol layer even at these low concentrations. Thisdisturbance is also visible in the capacitance scan as a distortion of the peak at thesame potential. At higher concentrations, a pair of pseudocapacitance peaks on theCV at approximately -0.15 V/SCE appear, along with a spike in the capacitancescans. These capacitance peaks are caused by potential driven changes in the orga-nization of the adsorbed layer. New peaks are an indication that the dye is causingsome disruption in the structure of the octadecanol layer.A comparison of the minimum capacitance values for each layer relates thegeneral quality of the adsorbed layer. The values for the minimum capacitance ofeach fluorophore concentration are overlaid with the ESP in Figure 3.2. A layerwith fewer defects and higher coverage will have a lower capacitance while ad-sorbed (for monolayers, between approximately -0.15 and +0.15 V/SCE). The pureoctadecanol unsurprisingly has the lowest minimum capacitance. The minimumcapacitance of the octadecanol monolayer containing 0.5 mol% of the fluorophoreis similar to the pure octadecanol value, reflecting the small amount of disruptioncaused by the low concentration. The 1 mol% and 5 mol% monolayers show ahigher minimum capacitance than the octadecanol reference, indicating that the923.3. Fluorescence - Floating Layers and Compressionsmonolayer structure is more significantly disrupted, containing more defects. The3 mol% monolayer shows the lowest minimum capacitance after the 0.5 mol%.Monolayers of the 8 mol% octadecanol mixture were also studied in this man-ner, however the quality of the monolayers were uneven, ranging from ?good qual-ity? layers similar to pure octadecanol to layers that showed more disruption thanthe 5 mol%, and a representative scan was unable to be selected. This behavioursuggests that at 8 mol%, there is a high degree of segregation in the layer, resultingin well-organized regions of nearly pure octadecanol, and highly disrupted regionsin the floating monolayer that are variously sampled on deposition onto the elec-trode.The trend in minimum capacitance does not follow the trend in fluorophoreconcentration. However, the trend does follow the trend in ESP (Figure 3.2) wherethe 3 mol% corresponds to a minimum in the ESP value.Thus, based on the studies of the electrochemical behaviour of the adsorbed oc-tadecanol monolayers, the 3 mol% concentration is recommended as the maximumfluorophore for future study.Since future studies on liposome incorporation would be performed with oc-tadecanol bilayers, measurements of bilayers with various concentrations of theBODIPY-C19-OH fluorophore were performed in the same manner. Figure 3.4shows typical CV and capacitance scans at concentrations from 0 to 3 mol% ofthe fluorophore in octadecanol. Again comparing the minimum capacitance valuesfor each layer, the 0.5 mol% solution shows the lowest capacitance value of thefluorophore-containing layers, and thus the least amount of disruption, followedby the 3 mol%. The 5 and 8 mol% solutions, already shown to be less stable inmonolayer studies, were not deemed appropriate for continued study of bilayerelectrochemistry.3.3 Fluorescence - Floating Layers and CompressionsMonolayers of octadecanol containing 1 mol% and 3 mol% of the BODIPY-C19-OH fluorophore were examined using a Langmuir trough equipped for in-situ flu-orescence measurements. Fluorescence imaging at the ESP of such layers havebeen reported previously [114, 162] as well as Brewster angle microscopy [40](reviewed in Sections 1.3 and 1.6.2), showing a variety of structures present inthe floating layer, both in pure octadecanol and in fluorophore-doped octadecanol933.3.Fluorescence-FloatingLayersandCompressionsFigure 3.3: Cyclic voltammograms (top) and differential capacitance scans (bottom) of octadecanol monolayers containing variousconcentrations of BODIPY-C19-OH on Au(111). Gray dotted lines are Au(111) with no octadecanol. From left to right: pureoctadecanol, 0.5 mol%, 1 mol%, 3 mol%, and 5 mol% BODIPY-C19-OH in octadecanol. CV measurements were acquired usinga scan rate of 20 mV/s. Capacitance measurements were made using a 5 mV/s scan rate, 5 mV RMS potential perturbation, and a25 Hz perturbation frequency. The increased capacitance (above the bare gold value) for 5 and 8 mol% is believed to be linked toincreased area caused by wetting of the sides of the electrode.943.3. Fluorescence - Floating Layers and CompressionsFigure 3.4: Cyclic voltammograms (top) and differential capacitance scans (bot-tom) of octadecanol bilayers containing various concentrations of BODIPY-C19-OH on Au(111). Gray dotted lines are Au(111) with no octadecanol. From leftto right: pure octadecanol, 0.5 mol%, 1 mol%, and 3 mol% BODIPY-C19-OH inoctadecanol. CV measurements were acquired using a scan rate of 20 mV/s. Ca-pacitance measurements were made using a 5 mV/s scan rate, 5 mV RMS potentialperturbation, and a 25 Hz perturbation frequency. The increased capacitance (abovethe bare gold value) is believed to be linked to increased area caused by wetting ofthe sides of the electrode as the meniscus was held lower due to reduced surfacetension in the presence of octadecanol.953.3. Fluorescence - Floating Layers and Compressionslayers. Study of the BODIPY-C19-OH containing monolayers will elucidate thestructure of the octadecanol monolayers used for adsorption onto the electrode sur-face, as well as the distribution of the fluorophore within the layer. The ability ofthe BODIPY fluorophore to form a dimer that emits a red-shifted fluorescence (seeSection 1.6.1) can be used as an indicator of the local concentration of the fluo-rophore within the layer. Regions of high fluorophore concentration should havea stronger dimer fluorescence signal. The higher concentration implies that thefluorophores are closer together, enabling both dimer formation and FRET energytransfer from the BODIPY monomers to create the dimer excited state. Thus, therelative monomer and dimer signal can be used as a rough indicator of the localconcentration of the BODIPY fluorophore.3.3.1 Experimental MethodsA Langmuir trough specially equipped for in-situ microscopy (Figure 2.1a) wasprepared according to the procedures in Section 2.3.1. An aliquot of the desiredoctadecanol solution (1 -3 ?L, less than the amount required for monolayer forma-tion at ESP) was deposited onto the air-electrolyte interface of the trough with thebarriers fully open. The solvent was allowed to evaporate and the layer to equili-brate until the surface pressure provided a stable value. Fluorescence images of thesurface were taken, first using the green emission filter (BODIPY monomer), thenthe red emission filter (BODIPY DII dimer). As the DI dimer is non-fluorescent(see Section 1.6.1), it is not detectable by this method. After images of the un-compressed layer were taken, the barriers were partially closed to achieve a filmpressure of between 25 and 35 mN/m, and further images taken with both the redand green fluorescence filters. This range of film pressure was chosen as it is nearthe ESP of the floating layers, but below the collapse point.Images were taken with a 10x objective, using an exposure time of 0.1 s andelectron multiplier gain of 200. In between images, the shutter was closed in orderto reduce photobleaching. The layer was slightly mobile on the surface, so althoughit was not possible to move the focal area around the interface to image differentareas, by waiting approximately 5 min between image pairs it was possible to viewnew regions of the floating monolayer with each acquisition.Because the filter cubes must be switched manually, there is a small time de-lay between green and red images, during which the floating monolayer may drift963.3. Fluorescence - Floating Layers and CompressionsFigure 3.5: Fluorescence images of a floating octadecanol monolayer containing1 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. Images taken withbarriers open; film pressure is less than 15 mN/m. Each vertical image pair is ofapproximately the same region of the interface, using the green emission filter (top)and red emission filter (bottom). Acquisition conditions: 10x objective, 0.1 secondexposure time, electron multiplier gain 200. Scale bar is 130 ?m.within the viewing area. For image analysis, the two images were first alignedin ImageJ using visible features and any non-overlapping area cropped out. Thealigned images were then processed with a Despeckle filter followed by a 2-pixelMedian filter. The aligned images were then divided (Red / Green) using the ImageCalculator function to provide the red:green fluorescence ratio.Regions within each image were identified as either ?background? - the lowerintensity, flat or slightly mottled regions typically found in most of the image; or?hotspots? - small regions of higher fluorescence intensity, and the red:green ratioanalyzed separately for each region type. Other structures, such as stripes or largebright regions, were identified but not included in the ratio analysis as they werenot present in all images analyzed.3.3.2 Uncompressed LayersIn the uncompressed, as-deposited floating monolayer, a variety of structures arevisible. In the images with 1 mol% of the BODIPY-C19-OH fluorophore in octade-973.3. Fluorescence - Floating Layers and CompressionsFigure 3.6: Fluorescence images of a floating octadecanol monolayer containing3 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. Images taken withbarriers open; film pressure is less than 15 mN/m. Each vertical image pair is ofapproximately the same region of the interface, using the green emission filter (top)and red emission filter (bottom). Acquisition conditions: 10x objective, 0.1X sec-ond exposure time, electron multiplier gain 200. Scale bar is 130 ?m.983.3. Fluorescence - Floating Layers and Compressionscanol (Figure 3.5), the observed structures include an overall mottled texture, in-clusions of differently-fluorescent regions, such as in the centre column, and brightstripes and hotspots, prominent in the rightmost image pair. In the small hotspotregions, the increased green fluorescence signal from the BODIPY monomers is ac-companied by an increase in red dimer fluorescence. This behaviour is as expectedfor a region enriched with the fluorophore, as the molecules present in higher con-centration should therefore be closer together and more likely to both form thedimer and become electronically excited through FRET with a nearby monomermolecule.A few regions (such as the ?blob? in the centre column in Figure 3.5 and thebottom corner of the leftmost column in Figure 3.6) show an increase in the greenfluorescence signal with no corresponding increase in the red intensity. It appearsthat in these regions the concentration of BODIPY-C19-OH is larger than surround-ing areas, but the fluorophores are positioned far enough apart that dimer formationis discouraged.Despite the differences in structure implied in the electrochemical results inSection 3.2, the structures seen in the 3 mol% layer are similar to those in the1 mol% layer. As monolayers studied electrochemically were formed from mono-layers at ESP, a study of more compressed monolayers will speak more directly tothe electrochemical results.3.3.3 Compressions of LayersAfter imaging at low surface pressure, the barriers of the Langmuir trough wereclosed until the surface pressure had increased to between 25 and 35 mN/m, avalue similar to the ESP, and below the collapse pressure of the layer. Althoughfor both 1 mol% and 3 mol% layers large areas of the same mottled texture seenin the uncompressed layers are common, in the 1 mol% layer the large bright re-gions seen in the centre and rightmost columns of Figure 3.7 are more commonthan in the 3 mol% layer. The presence of these regions agrees with the hypoth-esis based on the Langmuir isotherms that the dye is not mixing ideally with theoctadecanol in the monolayers. As seen in the uncompressed layers, these regionshave an increased green (monomer) signal but no increase in red (dimer) signal,so the molecular structure of these regions must be such that the fluorophores aresignificantly separated despite the increase in concentration.993.3. Fluorescence - Floating Layers and CompressionsFigure 3.7: Fluorescence images of a floating octadecanol monolayer containing1 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. Images taken withbarriers partially closed; film pressure is approximately 25 mN/m. Each verticalimage pair is of approximately the same region of the interface, using the greenemission filter (top) and red emission filter (bottom). Acquisition conditions: 10xobjective, 0.1 second exposure time, electron multiplier gain 200. Scale bar is130 ?m.1003.3. Fluorescence - Floating Layers and CompressionsFigure 3.8: Fluorescence images of a floating octadecanol monolayer containing3 mol% BODIPY-C19-OH fluorophore in a Langmuir trough. Images taken withbarriers partially closed; film pressure is approximately 25 mN/m. Each verticalimage pair is of approximately the same region of the interface, using the greenemission filter (top) and red emission filter (bottom). Acquisition conditions: 10xobjective, 0.1 second exposure time, electron multiplier gain 200. Scale bar is130 ?m.1013.3. Fluorescence - Floating Layers and CompressionsFeature Type Layer Status R/G Ratio(1% layers)R/G Ratio(3% layers)Background Uncompressed 0.392 0.438Compressed 0.538 0.550Hotspots Uncompressed 0.320 0.447Compressed 0.283 0.903Table 3.1: Average red:green (R/G) ratios for octadecanol monolayers containing1 mol% and 3 mol% BODIPY-C19-OH. Measurements are shown for the mottled?background? regions and for the small bright ?hotspot? regions in each imagetype. Images of the uncompressed layers were taken with trough barriers open;film pressure less than 13 mN/m. Images of the compressed layers were taken withbarriers partially closed at film pressures 25-35 mN/m. A selection of the imagesused for measurement are shown in Figures 3.5-3.8, however these ratio values arebased on images of 5-7 regions at each pressure and concentration.Comparing the ratio of the red:green (dimer:monomer) fluorescence intensitycan provide some information on the local variations in concentration of the BODIPY-C19-OH fluorophore within the monolayer. As the fluorophore concentration in-creases, the distance between the molecules should decrease, causing more dimerto be formed and increasing the likelihood of excitation by FRET with nearbymonomers. Thus, as the local concentration of the fluorophore increases, an in-crease in the red:green ratio should follow.The regions that were present in all images (the mottled ?background? andsmall hotspots) were analyzed and the average red:green ratios calculated for eachimage set. These ratios are presented in Table 3.1. In both the compressed and un-compressed image sets, the 3 mol% layer consistently has a higher red:green ratiothan the 1 mol% layer for each region, as expected for a layer of higher concen-tration. As the layers were compressed, the red:green ratio also increased for eachregion, with the exception of the 1 mol% hotspots. As the mean molecular areadecreases with compression, the molecules are closer together, and the amount ofdimer present therefore increases, increasing the red:green ratio. The decrease inratio for the 1 mol% hotspots might indicate that some of the areas identified as?hotspots? in the compressed layer might have a different structure than those ana-lyzed in the uncompressed or 3 mol% layers, perhaps similar to the intensely greenregions discussed above.These ratios do not provide a quantitative measure of the relative amounts of1023.4. In-situ Fluorescence with Electrochemistrymonomer and dimer present in the layer for several reasons. First, the DII dimerdoes not absorb strongly in the spectral region passed by the excitation filter, and isinstead excited by FRET from nearby monomer molecules. Thus, an increase in thedimer fluorescence must come at the cost of some monomer signal as the monomerexcited state is fed into the dimer excitation rather than directly fluorescing. Aswell, there may be dimer present that is unable to fluoresce due to a lack of nearbymonomer molecules. In addition to these chemical considerations, the sensitivity ofthe CCD used for image acquisition varies across the wavelength range, being moresensitive to red than green light. With these complicating factors, the intensities andratios from the fluorescence images can report only on the relative changes in thefluorophore state.3.4 In-situ Fluorescence with ElectrochemistryBased on the Langmuir isotherm, ESP, and floating layer fluorescence imaging,the 3 mol% concentration of BODIPY-C19-OH fluorophore appears to be the bestchoice for future study, providing the maximum fluorophore concentration andtherefore maximum fluorescence signal, while retaining the electrochemical char-acteristics of a well-organized octadecanol layer.For the studies investigating the interaction of liposomes with adsorbed octade-canol layers, a bilayer of octadecanol will be used. Therefore the fluorescence andelectrochemical characteristics of the BODIPY-C19-OH containing octadecanol bi-layers must be established.3.4.1 Experimental MethodsThe Au(111) electrode and in-situ fluorescence cell were cleaned and prepared asdescribed in Chapter 2. A bilayer of octadecanol was deposited by two sequentialLangmuir-Schaefer touches, the first ex-situ in a beaker containing a monolayer ofoctadecanol with 3 mol% BODIPY-C19-OH, and the second in the electrochemicalcell, of pure octadecanol. This is the same bilayer configuration that is used for allfollowing work studying liposome incorporation.The modified electrode was then scanned by cyclic voltammetry between +0.15and -0.15 V/SCE until successive scans overlapped, indicating that the sides of theelectrode in the hanging meniscus had dried, and the deposited layer was stable.1033.4. In-situ Fluorescence with ElectrochemistryThe initial capacitance was measured by a differential capacitance scan from +0.15to -0.15 V/SCE.In-situ fluorescence measurements were taken using the 50x objective and anexposure time of 2.5 sec and electron multiplier gain of 200, with images takenevery 5 sec. The potential was stepped by 25 mV with each image to simulatea 5 mV/sec scan rate. The electrode was first stepped from +0.15 V/SCE to -0.5 V/SCE and back to explore the fluorescence behaviour of the electrode duringlayer poration. This scan was repeated three times, twice with the green emissionfilter, and a third time with the red emission filter. The octadecanol bilayer was thendesorbed and readsorbed by cycling the potential from +0.15 V/SCE to -0.8 V/SCEand back. This scan was also repeated three times, twice with the green emissionfilter and a final time with the red emission filter.The triple cycle method was used in order to measure fluorescence scans withboth monomer (green) and dimer (red) fluorescence images that are able to bedirectly compared. As previously established (see Section 1.5.2) the virgin oc-tadecanol layer formed with the Langmuir-Schaefer method has different electro-chemical properties than the subsequent desorptions due to octadecanol moleculeschanging orientation. Thus, the first scan in these experiments was not used foranalysis, and only the second (green) and third (red) desorption scans were usedfor comparison.3.4.2 ResultsFluorescence and differential capacitance scans for the octadecanol bilayer areshown in Figure 3.9. The scans to -0.5 V/SCE - Figure 3.9a and b - show a smallchange in capacitance starting at approximately -0.4 V/SCE associated with theformation of defects in the adsorbed layer. However, there is no correspondingchange in fluorescence visible at these imaging conditions in either the green orred fluorescence signals. This is not unexpected as the change in layer structure issubtle. Since the octadecanol bilayer remains adsorbed at the electrode surface, thefluorescence remains significantly quenched.The capacitance scan produced by the second and third desorptions of the layer(Figure 3.9d) are similar to those already seen for the octadecanol and octadecanol-/BODIPY-C19-OH bilayers in Figure 3.4. Although the changes from the firstscan (not shown) to the second scan are large, the changes between the second1043.5. Conclusionsand subsequent scans are much smaller - this can be seen in the small shift incapacitance between the green (second) and red (third) capacitance scans.The fluorescence during desorption shows very little change in either the monomeror dimer fluorescence until approximately -0.6 V/SCE, correlating with the largeincrease in capacitance as the octadecanol layer is desorbed from the electrode sur-face. The fluorescence reaches a maximum at the minimum potential (-0.8 V/SCE),where the layer is fully desorbed. Fluorescence decreases steadily due to pho-tobleaching until the layer begins to readsorb at approximately -0.3 V/SCE. Theonset of readsorption is accompanied by a small bump in fluorescence, then a sharpdecrease to a minimum value as the layer becomes fully readsorbed. The fluores-cence continues to decrease on the scan from the positive limit back to 0 V/SCE,as does the capacitance, suggesting that the readsorption process is slow, and somechange in the layer structure continues even after the layer is nominally readsorbed.The dimer (red) fluorescence follows the same basic trend as the green fluo-rescence, although the absolute intensity is much lower, and the change in fluo-rescence during desorption is also smaller. The fact that there is still some sig-nal from the BODIPY dimers at desorption indicates that the fluorophores are notseparating from each other enough during the desorption process to break all ofthe dimers. However, interpreting the relative fluorescence between monomer anddimer is fraught with complication - in addition to those described in Section 3.3.3,the quenching characteristics for the monomer and dimer will differ based on theemission wavelength. Although there has been some work to suggest that the redfluorophore will be quenched at greater fluorophore-electrode separations than thegreen [114], the exact characteristics of the quenching remain unknown. For thesereasons, the dimer fluorescence is generally not considered, and future results willonly present the monomer fluorescence.3.5 ConclusionsBased on the compression isotherms and ESP measurements, increasing concentra-tion of BODIPY-C19-OH in the octadecanol monolayer causes significant disrup-tion to layer organization, and the floating layer appears to segregate into octadecanol-rich and fluorophore-rich regions. These observations are supported by fluores-cence measurements of the floating monolayer, where various structures formed inthe layer are clearly visible. The 3 mol% concentration appears to be the maxi-1053.5. ConclusionsFigure 3.9: In-situ fluorescence and differential capacitance measurements of abilayer of octadecanol containing BODIPY-C19-OH fluorophore. (a) and (b): Flu-orescence and capacitance of a preliminary scan to -0.5 V/SCE, where the layercontains defects but is not desorbed. (c) and (d): Fluorescence and capacitancedesorption scans of the same layer (to -0.8 V/SCE). All scans are the second andthird scans to their respective potential limits. Images acquired using 50x objec-tive, 2.5 sec exposure time and electron multiplier gain 200. The values for thered (dimer) fluorescence have been multiplied by 2 so that they are visible on thesame plot as the green. Images on the right are the maximum fluorescence imagesat desorption (-0.8 V/SCE) for monomer and dimer.1063.5. Conclusionsmum BODIPY-C19-OH concentration at which the structural properties (based onLangmuir isotherms) and electrochemical behaviour (in both mono- and bi-layers)remain essentially similar to pure octadecanol. Based on these findings, the octade-canol layers used for the liposome interaction studies will be the 3 mol% BODIPY-C19-OH concentration.107Chapter 4Electrochemical Studies ofLiposome Interaction withSolid-Supported OctadecanolBilayersIn the Introduction and Section 1.8, a method for forming a hybrid layer of an ad-sorbed surfactant bilayer and a phospholipid bilayer was proposed. In this method,surfactant layer is initially deposited onto an electrode surface. Defects are cre-ated in this layer, into which phospholipid liposomes may insert themselves andincorporate into the adsorbed layer. Figure 1.54 outlines the scheme graphically.This chapter describes the initial investigations into forming such a hybrid layer.As highlighted in Figure 1.55, reproduced here as Figure 4.1, there is a potentialwindow where an adsorbed octadecanol bilayer is expected to contain potential-induced defects, while a DOPC bilayer, if adsorbed, will form a stable adsorbedstate. Based on this observation, octadecanol has been selected as the model for theadsorbed surfactant layer to which liposomes of DOPC are introduced. Differentialcapacitance is used to monitor bulk changes in the layer properties. Although thisis a convenient method and does not require any special treatment of the electrodeor liposomes, capacitance alone is unable to determine the nature of any changesthat do occur, and can only provide average values representing the entire electrodesurface; no spatial information may be inferred.In this context, ?interaction? or ?incorporation? of liposomes into the existingoctadecanol bilayer is taken as a change in the capacitance behaviour - either achange in the capacitance value when measured at a constant potential, or as a shiftin the potential at which characteristic changes in the capacitance, such as desorp-tion, occur. It is possible that some interactions may take place to which capaci-108Chapter 4. Electrochemical Studies of Liposome Interaction...05101520253035Capacity (?F/cm2 )DOPC bilayerbare Au(111)05101520253035-1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2Capacity (?F/cm2 )E (V vs. SCE)Octadecanol bilayerbare Au(111)Figure 4.1: Differential capacitance scans during desorption of a DOPC bilayer(top frame) and octadecanol bilayer (bottom). The region highlighted in blue ex-tends from the onset of defect formation in octadecanol (-0.2 V) to the beginningof the first phase change in the DOPC bilayer (-0.4 V/SCE), representing the po-tential range in which liposomes are expected to be most effective at incorporatinginto a solid-supported octadecanol bilayer. Capacitance measurements were per-formed with a 5 mV/s scan rate, 5 mV RMS potential perturbation, and a 25 Hzperturbation frequency.1094.1. Electrochemistry of DOPC on Au(111)0510152025303540-1.200 -1.000 -0.800 -0.600 -0.400 -0.200 0.000 0.200Capacity (?F/cm2 )E (V vs. SCE)DOPCbare AuDepositionsFigure 4.2: Differential capacitance measurements of DOPC on Au(111). Gray,dashed line is the capacitance of unmodified Au(111). Red line is a desorption scan(from +0.15 to -1.1 V/SCE) of Au(111) exposed to DOPC liposomes at 0 V/SCE.Blue dots represent equilibrium capacitance values for Au(111) exposed to DOPCliposomes in solution at the potentials indicated. Multiple depositions at 0 V/SCEare included for comparison. Capacitance measurements were performed with a5 mV/s scan rate, 5 mV RMS potential perturbation, and a 25 Hz perturbationfrequency.tance is insensitive, for example liposomes that adsorb without bursting onto theouter surface of the octadecanol layer. As the incorporation of liposomes into theadsorbed octadecanol layer will be distinguishable by changes in capacitance, theseother interactions will be discounted in the current discussion, and other methods,such as in-situ fluorescence and atomic force microscopy, will be used to charac-terize the interactions.4.1 Electrochemistry of DOPC on Au(111)The behaviour of an Au(111) electrode exposed to liposomes of DOPC has beenpreviously documented in the literature[126, 129, 132] and summarized in Section1104.1. Electrochemistry of DOPC on Au(111)1.5.3. When a Au(111) electrode is placed in contact with a solution containingphospholipid liposomes, the liposomes spread onto the surface to form a bilayercoating the interface. The layer as-formed is not perfect, and is believed to containwater trapped in the bilayer and possible defects.At potentials more positive than -0.4 V/SCE, the layer is adsorbed onto theelectrode surface as a condensed bilayer phase. Potentials more negative than -0.4 V/SCE cause a thickening of the bilayer and a phase change, evidenced by thesharp rise in capacitance at these potentials, followed by an eventual separation ofthe bilayer from the electrode at -1.1 to -1.2 V/SCE. This behaviour is illustrated bythe red trace in Figure 4.2, showing the negative-going potential scan of a DOPC bi-layer formed by exposing the Au(111) electrode to a liposome-containing solutionwhile being held at 0V/SCE, where the water-covered electrode surface is nega-tively charged, as the pzc of Au(111) in 0.1 M NaF is approximately 0.330 V/SCE[108]. Note that at potentials more negative than -0.4 V/SCE, the capacitance ofthe DOPC-modified electrode is higher than that of the unmodified (bare) electrode.This is due to a combination of pseudocapacitance peaks increasing the apparentcapacitance (see Sections 1.5.1 and 1.5.3) and wetting of the electrode?s sides, in-creasing the exposed area.In order to characterize the ability of liposomes to interact with the electrodeat other potentials, a freshly flame-annealed Au(111) electrode was exposed toliposome-containing electrolyte for one hour while being held at various poten-tials (See Section 2.4.1 for procedures). At the end of the hour, an equilibriumvalue of capacitance was achieved, shown as the blue points in Figure 4.2. Severalrepetitions of the procedure at 0 V/SCE are presented for reference on the repro-ducibility of the depositions. Small changes in the effective electrode area causedby variations in argon flow, meniscus height, etc., as well as possible variations insurface coverage, contribute to the variability of the capacitance values reported.Since the variation in electrode area is the largest source of error in these measure-ments, the relative distribution of capacitance values of the replicates measured at0 V/SCE should be representative of those measured at other potentials.Although only one data point is presented, a similar range of responses can beexpected for depositions at all potentials, since the external variables affecting theelectrode area are similar across experiments.From Figure 4.2, it is clear that the liposomes are effectively adsorbing onto theelectrode surface at all potentials tested (0, -0.35, -0.45, and -0.8 V/SCE), produc-1114.2. Interaction of DOPC with Floating Octadecanol MonolayersDescriptionEquilibrium SurfacePressure (?, mN/m)Octadecanol with0.23 mol% DOPC 30?1Pure octadecanol 34?1Octadecanol with liposomes in solution 35-37Octadecanol with liposomes in solution,after layer re-formed with chloroform 32?1Table 4.1: Equilibrium surface pressure (ESP) of octadecanol and mixed octade-canol/DOPC a layer that is similar (within expected experimental error) to the capacitanceof the layer during a potential sweep of a previously formed bilayer. Based onthese results, it is shown that the liposomes are able to interact with the Au(111)electrode effectively within the entire potential range of octadecanol stability, froma completely adsorbed layer at 0 V/SCE to complete desorption of octadecanol at-0.8 V/SCE. Thus, the liposome-octadecanol combination is suitable for investiga-tion of the potential-controlled interaction of the layers.4.2 Interaction of DOPC with Floating OctadecanolMonolayersIn order to determine the degree of interaction between octadecanol and DOPC li-posomes in the absence of potential control, the equilibrium surface pressure (ESP)of octadecanol monolayers was monitored with exposure to DOPC and liposomes.Table 4.1 summarizes the data from these measurements. Pure octadecanol in theabsence of liposomes has a measured ESP of 34 mN/m. This layer, as describedin Section 1.3, is imperfect and contains some defects and domain boundaries thatmight be expected to interact with liposomes in the subphase. This hypothesis canbe tested by comparing the interaction of liposomes with the floating monolayerwith layers known to contain a mixture of octadecanol and DOPC.Mixed layers created by deposition of a octadecanol/DOPC solution have asubstantially lower surface pressure than pure octadecanol, at 30 mN/m. The0.23 mol% concentration used approximates a possible DOPC concentration inan octadecanol layer modified according to the scheme proposed in the Introduc-1124.2. Interaction of DOPC with Floating Octadecanol Monolayers10152025303540450 10 20 30 40 50 60 70 80 90 100 110? (mN/m)Time (min)add octadecanolinject liposomesre-form with chloroformFigure 4.3: Surface pressure (?) during exposure to liposomes. Pure octadecanolwas initially added and allowed to equilibrate. At approximately 28 minutes, li-posome suspension was injected into the subphase. After a further 40 minutes ofequilibration (t=75 min), chloroform was added to the floating octadecanol mono-layer and allowed to dry.1134.3. Interaction of DOPC Liposomes with Octadecanol on Au(111) at 0 Vtion. Therefore, incorporation of DOPC liposomes into an octadecanol monolayeris expected to lower the measured ESP.To determine this, an octadecanol layer was formed and allowed to stabilizeat the ESP, then DOPC liposomes were injected into the subphase at a concen-tration approximately the same as in the electrochemical cell (30 ?g DOPC/mLelectrolyte). Surface pressure was allowed to equilibrate for at least 40 minutes tomeasure any effect of liposome presence on the surface pressure. One dataset ofsurface pressure measurement is shown in Figure 4.3. In both replicates, the sur-face pressure increases slightly after addition of liposomes to the subphase. Thisis likely an artifact caused by the disruption in the floating octadecanol monolayercaused by pushing the syringe needle through the air-solution interface, rather thanan incorporation of liposomes into the layer. The volume of liposome solutionadded is less than 1% of the solution volume in the beaker and should not affect thesurface pressure measurement by altering the solution levels.Re-forming the octadecanol monolayer by addition of chloroform to the air-solution interface (as described in Section 2.4.4) will extract some of the chloroform-soluble DOPC from the aqueous phase into the organic phase, mixing it with thedissolved octadecanol. Re-forming octadecanol layers without liposomes in solu-tion results in no shift or a slight increase in ESP. Re-forming an octadecanol layerwhen DOPC liposomes are present in the subphase results in a decrease in ESP(from 35 to 32 mN/m), similar to the mixed monolayer measurements, suggestingthat DOPC from the liposomes can be introduced into the layer in this way.From these results it is clear that the liposomes do not interact in a substan-tial way with the floating octadecanol monolayer, despite the expected presence ofsome defects at the interface.4.3 Interaction of DOPC Liposomes with Octadecanol onAu(111) at 0 VAlthough liposomes may not interact strongly with a floating monolayer of octade-canol, additional defects may be formed in the monolayer upon deposition onto anelectrode surface that would allow liposomes to interact with the solid-supportedlayer interface. Deposition of a bilayer rather than a monolayer is expected to covermany of these defects, however lack of interaction is not guaranteed. In order toinvestigate the liposome interaction with the as-formed octadecanol bilayer, differ-1144.3. Interaction of DOPC Liposomes with Octadecanol on Au(111) at 0 VFigure 4.4: Differential capacitance of octadecanol bilayers at 0 V/SCE duringexposure to DOPC liposomes for 60 min. All layers were pre-selected as describedin Section 2.5.2 to have a similar initial capacitance. Capacitance measurementswere performed with a 5 mV/s scan rate, 5 mV RMS potential perturbation, and a25 Hz perturbation frequency.ential capacitance was monitored during the 60 minute mixing time after additionof liposomes to the electrolyte. Note that in the DOPC-only experiments describedin Section 4.1, 60 minutes was more than sufficient time for the liposomes to forma layer on the electrode surface with a stable capacitance value.Exposure of an octadecanol-modified Au(111) electrode to a liposome solu-tion generally produced two types of behaviour, as depicted in Figure 4.4. Themost common behaviour was simply no change in the capacitance value with time,within the typical variation of an octadecanol layer (traces 1A and 1B). As in theDOPC measurements described previously, variations in the capacitance can becaused by deposition differences, and by wetting and dewetting of the sides of theelectrode in the hanging meniscus. Rarely (3 out of over 27 datasets), a depositionwould result in the behaviour in trace 2 - namely, a marked increase of capaci-tance with time. Despite the pre-screening of layers to have an initial capacitanceless than 1.08 ?F/cm2(Section 2.5.2), some octadecanol bilayers are neverthelessformed as substantially defective, allowing the interaction of liposomes even at0 V/SCE. These layers were considered to be faulty depositions, and are excludedfrom further analysis.At 0 V/SCE, where the adsorbed octadecanol bilayer has few or small defects,liposomes of DOPC in solution therefore do not incorporate into or disrupt the ex-isting layer, as determined by differential capacitance. This does not exclude the1154.4. Effect of Poration Potential on Liposome - Octadecanol Interactionpossibility that liposomes interact with the layer in other ways to which capaci-tance is insensitive, for example adsorbing on top of the octadecanol layer withoutdisrupting it. Further study (described in Chapter 5) will address these questions.4.4 Effect of Poration Potential on Liposome -Octadecanol InteractionAfter establishing that DOPC liposomes in solution do not incorporate into octade-canol bilayers in the absence of substantial defects in the bilayer, electrochemicalmethods of artificially forming defects in the layer were attempted. Based on theelectrochemical behaviour of octadecanol (summarized in Section 1.5.2), applica-tion of a potential more negative than -0.4 V/SCE should result in formation ofdefects in the octadecanol bilayer. By application of a series of potential steps,the ability of liposomes to interact with the octadecanol bilayer was tested. Aftermeasurements at 0 V/SCE to characterize the unmodified bilayer, a ?poration po-tential? - the potential at which defects were formed in the bilayer - was applied.A schematic of possible defect structures for an adsorbed bilayer of octadecanol isshown in Figure 1.24e.4.4.1 Experimental DesignElectrochemical measurements were performed as outlined in Chapter 2, and differ-ential capacitance monitored during all stages. Briefly, after deposition and screen-ing of an octadecanol bilayer (inner leaflet: octadecanol with 3 mol% BODIPY-C19-OH, outer leaflet: 100% octadecanol), liposomes were injected directly intothe electrolyte and the potential held at 0 V/SCE for one hour to allow diffusion intosolution (liposome injection was omitted for control experiments). Then a seriesof potential steps was applied to the electrode to explore the facility of liposomeinteraction with the adsorbed octadecanol layer as it was perturbed by applicationof the potentials.In one variation, outlined in Figure 4.5a, after being held at 0 V/SCE for 15 min-utes to establish a baseline capacitance, the potential is scanned to -0.2 V/SCE andheld for 15 min before returning to 0 V/SCE for a further 15 min. For most ex-periments, a four-step profile was used (Figure 4.5b). In these profiles, after theinitial 15 minutes at 0 V/SCE, the potential was then scanned to a ?poration po-1164.4. Effect of Poration Potential on Liposome - Octadecanol Interaction(a) (b)Figure 4.5: Potential step profiles used in testing liposome incorporation. (4.5a)In experiments following 4.5b, the time at each potential varied from 1 to 45 min,and one of the three ?poration potentials? shown (-0.4, -0.6, and -0.8 V/SCE) waschosen.tential? which ranged from -0.4 V/SCE to -0.8 V/SCE, and held for a time varyingbetween 1 min and 15 min. After this time, a -0.2 V/SCE potential was applied for15 min, followed by 0 V/SCE for a final 15 min. Exceptions to this pattern werefor the -0.8 V/SCE poration potential, where the potential was not held but merelyscanned from 0 V to -0.8 V and then immediately to -0.2 V/SCE. For the 45 minuteresidence time at -0.4 V/SCE, all potential steps were held for 45 min rather than15 min.After application of the stepping potential profile, the potential was scannedto desorption (+0.15 V to -0.8 V/SCE) to further characterize the changes to themodified layer.4.4.2 Capacitance During Potential StepsThe capacitance data from a data series exploring the effect of an increasing pora-tion potential with a constant hold time of 15 minutes are summarized in Figure 4.6.When using only the -0.2 V/SCE potential (Figure 4.6a), where few if any defectsshould be created, very little change in the bilayer is observed. Both the layer ex-posed to liposomes in solution and the liposome-free control show a small increasein capacitance during the -0.2 V/SCE potential step, as expected from known ca-pacitance properties of the layer. Upon return to 0 V/SCE, both samples return tovery near the initial capacitance value, with little lasting change to the layer caused1174.4. Effect of Poration Potential on Liposome - Octadecanol InteractionFigure 4.6: Differential capacitance of octadecanol bilayers on Au(111) with andwithout liposomes in solution during application of potential steps. Inset plots showthe potential profile applied in each experiment. (a) Application of -0.2 V/SCE po-tential step only. (b) Application of -0.4 V for 15 min, followed by -0.2 V/SCEfor 15 min. (c) Application of -0.6 V for 15 min, followed by -0.2 V/SCE for15 min. (d) Application of -0.8 V, followed by -0.2 V/SCE for 15 min. Capaci-tance measurements were performed with a 5 mV/s scan rate, 5 mV RMS potentialperturbation, and a 25 Hz perturbation frequency.1184.4. Effect of Poration Potential on Liposome - Octadecanol InteractionFigure 4.7: Replicate measurements of octadecanol capacitance with and withoutliposomes in solution during application of the (a) -0.6 V and (b) -0.4 V/SCE po-ration potential (as in Figure 4.6c). Solid lines: Electrochemistry in standard cellwith 30 ?g/mL DOPC, performed with a 5 mV RMS potential perturbation, anda 25 Hz perturbation frequency.. Dashed lines: Electrochemistry in microscopecell with 45 ?g/mL DOPC, performed with a 5 mV RMS potential perturbation,and a 200 Hz perturbation frequency. A zoomed view of the capacitance after thepotential perturbations is shown in figure (c) for -0.6 V and (d) for -0.4 the excursion to the negative potential. It is likely that any defects formed at-0.2 V/SCE are too small or too few to facilitate substantial liposome-octadecanolinteraction.Addition of a step to -0.4 V/SCE produces more dramatic results (Figure 4.6b).The capacitance at -0.4 and -0.2 V/SCE is increased, as in the previous experiment,due to the presence of defects in the adsorbed layer and increase in the averagedielectric value caused by water at the electrode surface. However, the layer in thepresence of liposomes experiences a larger relative change in the capacitance at -0.4 V/SCE. The cause of the ?jump? in capacitance at approximately 28 minutes isunknown, and did not occur in other measurements. As well, rather than returningto a value similar to the initial capacitance upon return to 0 V/SCE, the liposome-exposed layer maintains a distinctly higher final capacitance value. This increasein capacitance signals a decrease in the organization of the octadecanol layer, pre-sumably caused by incorporation of liposome material. The addition of a step to-0.4 V/SCE results in greater liposome interaction with the adsorbed octadecanolbilayer, likely due to increased defect formation at this potential.If the poration potential is extended to -0.6 V/SCE, the change in capacitance at1194.4. Effect of Poration Potential on Liposome - Octadecanol Interactionthe poration potential is correspondingly larger, however upon return to 0 V/SCE,little or no change is seen from the initial capacitance values. A repetition ofthe measurement in the smaller microscope electrochemical cell (using a slightlyhigher liposome concentration - 45 rather than 30 ?g/mL DOPC) is shown alongwith the original data in Figure 4.7. Both of the liposome-exposed layers in themicroscope cell (gray, dashed lines in Figure 4.7) showed a larger change in capac-itance than the standard cell experiments (solid lines), which is easily attributableto the increase in liposome concentration in this cell. More importantly, the con-trol layers (black lines) show markedly different responses to the same potentialperturbation, especially during the poration step (15 to 30 min), where the secondexperiment (dashed line), although it started at a similar value to the first (solidline), experienced a much greater increase in capacitance, and higher capacitancevalues remain thereafter. Contrast this variation to the typical behaviour of the -0.4 V/SCE potential perturbation (Figure 4.7b), where the control layers, despitebeing measured in two different cells again, exhibit almost identical capacitancebehaviour.The difference is seen even more clearly in the expanded plots of Figure 4.7cand d. With a -0.4 V/SCE poration potential, both control experiments have verysimilar capacitances (which are again similar to the initial capacitance values seenin Figure 4.7b). With the -0.6 V/SCE poration potential, although both controllayers again started with similar capacitances, there is a much greater difference infinal capacitance from run to run. Clearly, the -0.6 V/SCE potential profile suffersfrom a lack of reproducibility.The source of this variation is most likely small changes in layer formationconditions, resulting in a shift in the desorption potential. The onset of octadecanoldesorption is typically near 0.6 V/SCE, so a small shift in the desorption potentialwill result in widely varying conditions at the electrode surface. The -0.4 V/SCEporation potential is further from such large-scale transitions, and is less affectedby small variations in initial octadecanol conditions. Therefore, the -0.4 V/SCEpotential profile is recommended over the -0.6 V/SCE profile, as the -0.6 V/SCEprofile suffers from reproducibility problems that are not easily detected at the startof experiments.The final potential tested was -0.8 V/SCE, at which the octadecanol layer iscompletely desorbed from the electrode surface. Although previous research hasshown that the octadecanol layer does not travel far from the interface while des-1204.4. Effect of Poration Potential on Liposome - Octadecanol Interactionorbed [112, 115, 183?185], this state is expected to be less stable with time thanthe adsorbed states. Rather than holding the potential at -0.8 V/SCE for 15 min-utes, it was scanned out and immediately back to -0.2 V/SCE. Although the time at-0.8 V/SCE is relatively short (a few seconds), based on the previously describedcapacitance behaviour, the layer should not readsorb onto the electrode surface im-mediately, staying substantially desorbed until the potential reaches approximately-0.4 V/SCE. Thus, this potential profile results in the liposome solution being ex-posed to the interface with a desorbed octadecanol bilayer for several minutes over-all. The capacitance behaviour during this trial (Figure 4.6d) shows the expecteddramatic increase in capacitance as the layer is desorbed at the most negative poten-tials. Upon return to 0 V/SCE, the liposome-exposed layer does show an increasein capacitance over the control layer, indicating that liposomes indeed are able tointeract with the octadecanol layer through the process of desorption and readsorp-tion. However, the overall change is smaller than that of the -0.4 V/SCE porationpotential, suggesting that less liposome material has been incorporated into theoctadecanol bilayer. This may be due in part to the shorter time spent at the po-ration potential, but increasing the residence time at desorption is inadvisable asthe octadecanol layer is only very loosely associated with the electrode, and willeventually diffuse away.By observing the capacitance behaviour during the poration and healing of theoctadecanol layer, it is shown that the application of a poration potential is neces-sary in order to achieve liposome interaction with the octadecanol layer. As well,the choice of potential is important in order to achieve reproducibility as well asquality of liposome interaction.4.4.3 Desorption of the Modified BilayersFurther characterization of the modified octadecanol layers was done by desorptionof the modified layer while monitoring changes in capacitance behaviour. Desorp-tion scans of the layers formed during the procedures shown in Figure 4.6 are shownin Figure 4.8. Generally, a shift in the desorption potential to more positive valuesindicates an increase in layer disorder, allowing the layer to desorb at lower en-ergy. However, this information is convoluted with the fact that it is impossible toprevent liposomes from interacting with the layer as it becomes porated before des-orption. Therefore, even layers that have little liposome incorporation prior to the1214.4. Effect of Poration Potential on Liposome - Octadecanol InteractionFigure 4.8: Capacitance during a potential sweep to desorption (+0.15 to -0.8 V/SCE) of octadecanol layers subjected to the potential profiles in Figure 4.6.Data is from the first potential sweep taken immediately after application of the po-ration potential profile. (a) Layers after application of a -0.2 V/SCE potential. (b)Layers after application of a -0.4 V followed by a -0.2 V/SCE potential. (c) Layersafter application of a -0.6 V followed by a -0.2 V/SCE potential. (d) Layers afterapplication of a -0.8 V followed by a -0.2 V potential. Capacitance measurementswere performed with a 5 mV/s scan rate, 5 mV RMS potential perturbation, and a25 Hz perturbation frequency.1224.5. Effect of Poration Time on Liposome - Octadecanol Interactionstart of the desorption scan would be expected to have some shift in the desorptionpotential caused by incorporation during the scan.Both of the potential profiles that showed little change during the applicationof poration potentials (-0.2 V and -0.8 V, Figure 4.8a and d respectively) show onlya small shift in desorption potential as well, confirming that little change occurredin the octadecanol layer during the application of the potential steps. Both (c) and(d) in Figure 4.8 (-0.4 and -0.6 V/SCE poration) show more significant shifts in thedesorption behaviour, corresponding with the larger changes observed during thepotential steps. In both cases, the onset of desorption, identified by the steep rise incapacitance, begins at more positive potentials for the layers exposed to liposomes.As there is very little variation run-to-run in the octadecanol layers not exposedto liposomes, these changes can be attributed entirely to the effects of liposomesinteracting with the octadecanol layer. Both the -0.4 and -0.6 V poration potentialsshow a similar ability to facilitate liposome interaction with the layer, as judgedby the relative shift in desorption potential - for both, beginning at approximately-0.45 V/SCE compared to -0.65 V/SCE for the unmodified octadecanol.Desorption scans of the liposome-modified and unmodified octadecanol bilay-ers show that potentials at which the octadecanol layers are adsorbed but experiencedisruption are effective in allowing liposomes to interact with the layer, while po-tentials at which the octadecanol is completely desorbed or weakly porated are lesseffective, confirming observations made during the application of the poration po-tentials. The -0.4 V/SCE poration potential is judged to be the most effective ofthose tested, as it allows significant interaction of liposomes with the octadecanolbilayer while avoiding the variability of the -0.6 V/SCE potentials. Thus, the -0.4 V/SCE potential was chosen for further investigation on the length of porationtimes.4.5 Effect of Poration Time on Liposome - OctadecanolInteractionIn order to determine the effect of the duration of the poration potential on thedegree of liposome interaction with the layer, a poration potential of -0.4 V/SCE(the best-performing potential as determined previously) was used with a variety ofresidence times.1234.5. Effect of Poration Time on Liposome - Octadecanol InteractionFigure 4.9: Differential capacitance of octadecanol bilayers on Au(111) with andwithout liposomes in solution during application of -0.4 V/SCE for various times.Inset plots show the potential profile applied in each experiment. (a) Applicationof -0.4 V/SCE for 1 min. (b) Application of -0.4 V for 15 min. Capacitance mea-surements were performed with a 5 mV RMS potential perturbation and a 25 Hzperturbation frequency.4.5.1 Experimental DesignThe same type of potential step pattern as described in Section 4.4 was applied,all with a minimum potential of -0.4 V/SCE. Generally, as in the previous experi-ments, the potential was held at 0 V/SCE for 15 minutes, then -0.4 V/SCE for therequired time (1 or 15 minutes), then -0.2 V and 0 V/SCE for 15 minutes each. Thelayers were then scanned to desorption to further characterize the changes made tothe layers. An experiment was also performed with the same general parameters,except all the potential steps were held for 45 minutes. Since the octadecanol isstably adsorbed at both 0 and -0.2 V/SCE, the additional time at these potentialsshould not affect the measurements, and these results may be compared directly tothe other, shorter durations.4.5.2 Capacitance During Application of Poration PotentialsAs previously, differential capacitance was monitored during the application of thepotential steps used to control the poration of the octadecanol bilayer. Figure 4.9shows the results of these experiments. Both 1 min and 15 min residence timesat -0.4 V/SCE were successful in allowing liposomes to interact with the octade-canol layer. Based on the relative increase in capacitance after application of thepotential steps, the 1 min residence time appears to be more effective at allowingliposome interaction, but the difference is small compared to the variability of the1244.5. Effect of Poration Time on Liposome - Octadecanol InteractionFigure 4.10: Capacitance during a potential sweep to desorption (+0.15 to -0.8 V/SCE) of octadecanol layers subjected to the potential profiles in Figure 4.9.Data is from the first potential sweep taken immediately after application of a po-tential step profile holding -0.4 V/SCE for (a) 1 minute (b) 15 minutes (solid line)and 45 minutes (dashed line). Capacitance measurements were performed witha 5 mV/s scan rate, 5 mV RMS potential perturbation, and a 25 Hz perturbationfrequency.experiments.This finding is interesting, but not entirely surprising - the potential steps areonly applied after the liposomes in the electrolyte have been allowed to diffuse foran hour, so the density of liposomes near the electrode surface is initially simi-lar for both experiments. It is therefore possible that if the octadecanol layer isnot altered during the residence time at -0.4 V/SCE (after the initial equilibrationof the layer) then the liposomes may achieve the maximum amount of interactionrelatively quickly, and further time at the poration potential does not enhance theinteraction . Rather, it is limited by liposome density at the surface. These hypothe-ses will be investigated further in Chapter 5, when fluorescence imaging is used tomonitor the changes during the application of these potentials.4.5.3 Desorption of the Modified LayersAfter application of the potential step profile, the modified octadecanol layers weresubjected to a potential scan to desorption (0.15 to -0.8 V/SCE) while measuringcapacitance, shown in Figure 4.10. All exposed layers show evidence of modi-fication by incorporation of liposomes, characterized by a shift in the desorptionpotential to be positive of -0.65 V/SCE.The 1 minute poration time (Figure 4.10a) showed a large shift in desorptionpotential. Longer poration times of 15 and 45 minutes (Figure 4.10b) also showedsome shift in the desorption potential, but increasing the time spent at -0.4 V/SCE1254.6. Conclusionsdoes not appear to bring a corresponding increase in the amount of change inducedin the adsorbed layer. Therefore, future studies of the layers will be confined toporation times of 15 minutes or less, as additional time provides no advantage tothe experiment.4.6 ConclusionsThe interaction of liposomes with a surface-supported octadecanol bilayer has beendemonstrated, using capacitance measurements. Application of a potential suffi-cient to cause disruption to the octadecanol bilayer is required in order to achievesignificant interaction of the liposomes with the octadecanol layer. In the absenceof sufficient defects, liposomes do not interact with the octadecanol layer in ameasurable way. Complete desorption of the octadecanol layer does not enhanceliposome-octadecanol interaction.The residence time at the poration potential does not appear to have a strongeffect on the degree of interaction of the liposomes with the octadecanol layer,although shorter times do appear to have a slightly larger degree of change. Overall,the optimum conditions for facilitating liposome incorporation into the octadecanolbilayer are application of a poration potential of -0.4 V/SCE for 1 or 15 min.Although the general properties of the interaction can be observed, the capac-itance measurements alone cannot provide information on how the liposomes areinteracting with the octadecanol layer, or whether the interactions are localized orheterogeneous across the electrode surface. These questions will be investigatedfirst using in-situ fluorescence methods, discussed in the next chapter.126Chapter 5In-situ Fluorescence Studies ofLiposome Interaction withSolid-Supported OctadecanolBilayersThe capacitance results presented in Chapter 4 provide convincing evidence thatliposome interaction with an adsorbed octadecanol bilayer can be initiated by ap-plication of an electrical potential, but electrochemical methods alone provide onlyaverage information on the state of the intervace. In order to learn about the na-ture of the interaction between liposomes and the adsorbed octadecanol layer, orgain any information on spatial characteristics of the interactions across the elec-trode surface, electrochemical measurements must be combined with other in-situmethods to probe the interface.Presented in this chapter are the observations of the interaction of liposomeswith adsorbed octadecanol using the combination of electrochemistry and in-situfluorescence microscopy. By the inclusion of a small amount of a fluorophoreinto the adsorbed octadecanol layer, information on the adsorption/desorption stateof the octadecanol layer can be obtained. The potential-dependent fluorescencebehaviour of monolayers of octadecanol is already well characterized [113, 114,183?185], and some observations of octadecanol bilayers [115] and DOPC mono-and bilayers [69, 113] using fluorescence have been published, and are summarizedin Section 1.6.2.The liposomes themselves contain no fluorophore, but as they interact withthe adsorbed octadecanol layer, some of the fluorophore contained in the adsorbedlayer may diffuse into the liposome walls. If these fluorophores are located in areasthat are now raised further from the electrode surface than the adsorbed layer (as127Chapter 5. In-situ Fluorescence Studies of Liposome Interaction...Figure 5.1: Cartoon illustration of a possible mechanism of increasing fluorescenceduring liposome incorporation. The initially non-fluorescent liposome leaches flu-orophore from the adsorbed ctadecanol layer, where its fluorescence is mostlyquenched. Fluorophore dissolved into the liposome structure is lifted away fromthe electrode surface, making it less quenched. In this situation, the raised hemi-liposome structure will be more fluorescent than the octadecanol background, evenif only a small amount of fluorophore is incorporated into the liposome. The blackbox at the octadecanol-DOPC interface indicates that the exact structure of the in-terfaced layers is not the hemi-liposome structure depicted in Figure 5.1), the quenching will be lessand the fluorescence greater in these regions. Since the relation between separationand transfer of energy from the fluorophore excited state is exponential (see Section1.6.1), even a small difference in height could result in a measurable difference influorescence intensity. Desorption of the layer, forcing all material to move fartherfrom the electrode surface, will enhance these differences.The structure of the fluorescent layer will be heavily influenced by the initiallayer structure and the particular area being imaged. As demonstrated in Chapter 3,the octadecanol/BODIPY-C19-OH layers are non-homogeneous at deposition andthis will influence the fluorescence intensity and structure of the layers under in-vestigation. Thus, although individual characteristics of the layers formed may notbe directly comparable, the trends in layer behaviors should be consistent betweenexperiments. Increases in fluorescence may therefore be interpreted as either layerdesorption, which moves octadecanol and fluorophore away from the electrode sur-face, or liposome incorporation, depending on the context of the experiment.1285.1. Experimental Methods5.1 Experimental MethodsThe electrochemical procedures used were as described in Sections 2.5 and 4.4.1.Octadecanol bilayers were formed by depositing first a monolayer containing 3 mol%of the BODIPY-C19-OH fluorophore, then depositing a layer of pure octadecanol.Deposition of the fluorophore-containing layer first (and thus nearest the electrodesurface) offers several advantages. While adsorbed, fluorescence from this layerwill be nearly completely quenched, reducing the background fluorescence as muchas possible.After deposition according to the procedures in Section 2.4.3, the capacitanceof each layer was measured and confirmed to be less than the maximum acceptablevalue of 1.08 ?F/cm2. Potential step profiles were applied to the layer, as describedin Section 4.4.1, followed by a potential scan to desorption (-0.8 V/SCE) and backto 0 V/SCE.Fluorescence images of the electrode surface were taken using the microscopeassembly described in Section 2.5.4. In order to maximize the available signal ofthe mostly quenched fluorescence in the adsorbed layer, images taken during the0 V/SCE hold time for liposome diffusion and during the application of potentialsteps were made using the green emission filter exclusively as the green monomerfluorescence is more intense than that of the dimer. Interpretation of dimer fluores-cence is complex, as established in Chapter 3, so analysis of the red dimer fluores-cence was not performed on these layers. During potential steps and scans, imageswere continuously acquired every 5 sec. During the initial hour at 0 V/SCE, imageswere acquired every 1-5 min in order to reduce the influence of photobleaching.Exposure time for the 60 min hold at 0 mV/SCE was 5 s, with an electron multi-plier gain of 400. Images during the potential stepping experiments were acquiredwith either a 2.5 s exposure and electron multiplier gain of 200, or a 5 s expo-sure with an electron multiplier gain of 400. For comparison, these images wereconverted to an equivalent kilocounts per second by use of an empirically deter-mined calibration factor (2.5 for the shorter exposure, 7.5 for the longer exposure).Derivation of the calibration factors is described in Appendix B.1295.1. Experimental MethodsFigure 5.2: Average fluorescence intensity and differential capacitance measure-ments for layers held at 0 V/SCE. Black lines are control experiments with noliposomes. Blue and red lines have liposomes in the electrolyte. Blue lines showa small change in fluorescence with time; red trace shows a larger fluorescencechange despite similar capacitance behaviour.1305.2. Fluorescence at 0 V5.2 Fluorescence at 0 VAfter depositing the octadecanol bilayer and adding liposomes to the electrolyte,the system was allowed to equilibrate for an hour to allow the liposomes to dif-fuse through the solution. This is also an opportunity to observe the interactionof the liposomes with the as-deposited octadecanol layer in the absence of poten-tial induced defects. In the standard electrochemical cell, little change in capac-itance was noted during this wait time at 0 V/SCE (Section 4.3). In the micro-scope cell, although the liposome concentration in the electrolyte was higher (45?g DOPC/mL electrolyte rather than 30 ?g DOPC/mL electrolyte), the capaci-tance values over the 60 minutes still show little change with time, as seen in Figure5.2b. Two control experiments are included to show typical octadecanol behaviourin the absence of liposomes; these show a similar stability.Despite the similar capacitance values, there is large variation in the initial flu-orescence intensity for the layers imaged. As outlined in Chapter 3, the floatingmonolayer of octadecanol with BODIPY-C19-OH used for deposition has a varietyof structures and fluorescence intensities that are transferred to the electrode surfaceon deposition. These differences in intensity are caused by variations in the localconcentration of fluorophore in the layer, resulting in regions that are dark - devoidof fluorophore - and intense fluorescence in fluorophore-rich regions. The elec-trode surface, at 0.26 cm2, will encompass several of these regions for any givendeposition while the fluorescence images, at 2.2?10?4 cm2, reflect only a smallportion of the surface. Thus, while the capacitance measurements reflect the aver-age value across the electrode surface, encompassing all regions, the fluorescenceimages reflect only the changes in the specific visible area, and the initial fluores-cence intensity depends on the type of region visible. It is probable that all layersdescribed here have these regions of fluorescent behaviour, indistinguishable by ca-pacitance, but obvious to fluorescence imaging. Because the fluorophore is mostlyquenched when the layer is deposited, little information is available on the layerstructure when choosing an imaging area, and the area imaged is essentially left tochance.Fluorescence measurements during the 60 minute waiting time reveal that de-spite the lack of measurable capacitance response, there is some change occurringin the layer while holding at 0 V/SCE. Little photobleaching is seen, likely becausethe layer is adsorbed onto the electrode and fluorescence is quenched, as well as the1315.2. Fluorescence at 0 Vreduced duty cycle used during imaging (one 5 sec image each 1 to 5 min). In thecontrol layers (black lines in Figure 5.2), the fluorescence intensity either remainsconstant or increases slightly over the wait time. This small increase - less than10% of the initial fluorescence value - is explained by a possible reorganization ofthe adsorbed bilayer such that fluorophore from the inner leaflet migrates to theouter leaflet, moving farther from the electrode surface and increasing fluorescenceintensity.When liposomes are added to the solution, they could interact with any defectspresent by slightly perturbing the octadecanol layer or leaching fluorophore into thephospholipid bilayer, which would result in a change in fluorescence dependent onthe structure of the region imaged. Both mechanisms of interaction would result inan increase in fluorescence by moving fluorophore away from the electrode surface,reducing the amount of quenching. The mechanisms should be distinguishable bycomparing the change in capacitance - a change in capacitance accompanying anincrease in fluorescence would indicate that the liposomes are incorporating intothe layer through existing defects, while a steady capacitance favors the leachingmechanism. In fact, in the majority of layers tested (about 60%) a modest increasein fluorescence of 10-15% of the initial fluorescence value is observed, similar inmagnitude to that seen in the control layers (blue lines in Figure 5.2). There is nochange in capacitance associated with this increase in fluorescence, indicating thatthe liposomes are not interacting significantly with the adsorbed octadecanol layerunder these conditions. In a few cases (about 20% of all experiments, representedby the red line in Figure 5.2), there was a larger increase in fluorescence, again notaccompanied by a change in capacitance. Analysis of these images shows eithera general increase in fluorescence across the image, or a change in fluorescenceconfined to the creation of small intensely fluorescent regions. In the latter case,these layers were abandoned as defective, as the fluorescence during subsequentsteps was substantially different than in the remainder of the experiments. Analysiswas also not performed on the small fraction of experiments where both the fluo-rescence and capacitance increased during the waiting time, clearly indicating thatliposomes were able to interact with existing defects in the octadecanol layer.Based on the observations of the octadecanol layers with and without lipo-somes before any potential perturbation was applied, it is clear that in the majorityof depositions, the liposomes are unable to incorporate on a large scale into theoctadecanol bilayer. There is some reorganization of the octadecanol bilayer that1325.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minresults in a small fluorescence increase. In a small fraction of experiments, suf-ficient defects were present at deposition to allow liposome interaction, howeverthese layers were excluded from further analysis.5.3 In-Situ Fluorescence Imaging: Poration at-0.4 V/SCE for 1 minIn order to investigate the interaction of DOPC liposomes with the adsorbed octade-canol bilayer, in-situ fluorescence images were taken during application of potentialprofiles similar to those used in Section 4.5. As the electrochemical measurementssuggested, the application of a -0.4 V/SCE potential for 1 min or 15 min facilitatedthe greatest interaction with the octadecanol layer; these experiments were repeatedwith the addition of in-situ fluorescence measurements.5.3.1 Fluorescence During Potential StepsThe capacitance and average fluorescence intensity profiles for layers exposed toa poration potential profile at -0.4 V/SCE for 1 min, identical to that used in theelectrochemical experiments of Figure 4.9a, are shown here in Figure 5.3. Onesample of a control experiment (black lines) and two experiments with liposomesin solution (blue and green lines) are included. The regions identified as hotspotsat 0 V/SCE are analyzed separately and shown as dashed lines. The capacitance inthe 15 min spent at 0 V/SCE is similar for all three examples. The effects of photo-bleaching are stronger in these data sets compared to the previous measurements at0 V/SCE as the imaging is done more frequently - here, one image per 5 sec as op-posed to every 1 to 5 min during the 60 min diffusion time. The fluorescence of thehotspots decreases more rapidly than the remainder of the image. As these are be-lieved to be multilayer regions that will be less efficiently quenched, this behaviourfits as the less quenched regions would be more subject to photodegredation.After changing the potential to -0.4 V/SCE, there is a significant increase incapacitance in the adsorbed layers caused by a change in the layer structure, similarto the increase in capacitance seen in the electrochemistry experiments in Section4.5. With no liposomes in solution, this change is small, and after returning thepotential to 0 V/SCE, the final capacitance value is only slightly increased from theinitial value at 0 V/SCE. The change in fluorescence during these changes was also1335.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minvery small, consisting mostly of photobleaching with time. Very little potentialdependence of the fluorescence is seen, with only a slight increase marking thechange im layer organization at -0.4 V/SCE.With liposomes in solution, both the capacitance and the fluorescence showan enhanced response to the changes in potential. Again, there is an increase incapacitance associated with the excursion to -0.4 V/SCE, however upon return to0 V/SCE, both layers exposed to liposomes retain a higher capacitance value thanthe initial capacitance at 0 V/SCE. The fluorescence response for these two layersdiffer significantly, illustrating that the fluorescence response observed can dependon the region used for analysis. In both layers, there is a small but significant in-crease in the overall fluorescence at -0.4 V/SCE, indicating the quick interactionof the liposomes with the adsorbed layer - the scan from 0 to -0.4 V/SCE takesonly 20 s, and the increase in fluorescence is visible even before the final potentialis reached. After changing the potential to -0.2 V/SCE, there is a sharp decreasein fluorescence athough the overall intensity continues to increase slowly in bothcases. On returning to 0 V/SCE, there is again an immediate decrease in fluores-cence intensity, and a continued increase with time, although the first experiment(green lines) shows a much smaller increase during this time. These changes influorescence suggest that the liposomes have disrupted the adsorbed layer in sucha way as to increase the separation of fluorophore from the electrode surface. Ithas been shown [113] that changes of 10-20 nm in this separation can result in asignificant increase in the fluorescence due to decreased quenching efficiency.The changes seen during poration are small and indicate that either there is nota large degree of liposome incorporation overall into the adsorbed layer or that fewfluorophores from the adsorbed layer have diffused into the adsorbed lipid regionsat the end of the measurement period. More detail on the interaction of the lipo-somes and the adsorbed layer can be gleaned from an analysis of the structure ofthe fluorescence images. A selection of fluorescence images acquired for each ofthe three experiments discussed here is presented in Figure 5.4. The changes influorescence are not uniform across the imaging area, but are localized in partic-ular regions, suggesting that the layer may have specific structures that facilitateliposome incorporation and interaction. These can be seen for example in the firstliposome exposure experiment (top row of Figure 5.4), where at the end of the -0.4 V/SCE poration (image C at -0.4 V/SCE and D at -0.2 V/SCE) new localizedspots of increased fluorescence are visible that were not included in the initially1345.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minidentified hotspots region. The effect is less visible in the second experiment as ageneral increase in fluorescence of the layer makes identification of new spots moredifficult. The control layer shows little change in fluorescence structure aside fromphotobleaching, either within the hotspots or in the surrounding regions.The increase in fluorescence clearly shows that the adsorbed layer is changingin its structure and supports the interaction of liposomes with the adsorbed layerbecause these changes are not soleley due to the potential perturbation of the ad-sorbed octadecanol layer. The increase in fluorescence is also not located aroundthe regions initially identified as hotspots, which suggests that the defects createdthrough the potential perturbation are not near the regions of high fluorescenceoutlined even though these regions would be considered to be a nonideally orga-nized region or a possible defect. Although some detail of this interaction is visibleduring the poration of the images, analysis is hampered by the relatively low flu-orescence as the adsorbed layer?s fluorescence is mostly quenched. Desorption ofthe layer from the electrode surface will highlight changes in the layer and allowgreater contrast between regions of the layer as the fluorescence intensity increases.5.3.2 Fluorescence During Desorption of the Modified LayersAfter modification with the potential step profile, the layers were desorbed by ap-plication of a potential scan from +0.15 to -0.8 V/SCE while measuring capaci-tance and fluorescence images. As demonstrated in the electrochemical measure-ments (Chapter 4), the interaction of liposomes with the adsorbed layer influencesthe potential-induced desorption, moving the onset of desorption to less-negativevalues. Without liposomes, the control layer behaves consistently with previousobservations; the black lines in Figure 5.5 show that the increase in fluorescenceintensity begins at approximately the same potential as the sharp increase in capac-itance, signaling desorption of the layer. As the capacitance reaches its maximumvalue, so does the fluorescence intensity, reaching a value nearly double the ini-tial intensity of the initial adsorbed layer. The regions identified in the initiallyformed layer as hotspots have a similar behaviour to the bulk of the layer. Thestrong increase in fluorescence occurs as the adsorbed layer is moved away fromthe electrode surface, reducing the degree of quenching of the fluorophores. Theintensity of the fluorescence signal is also influenced by the amount of dye initiallypresent in the adsorbed layer, the distribution of fluorophore normal to the electrode1355.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minFigure 5.3: Spectroelectrochemical behaviour of adsorbed octadecanol bilayerswith and without liposomes in solution during application of a potential profilewith a 1 min poration step. (a) Fluorescence intensity as average grayscale valuesfor hotspot regions (dotted lines) and the remainder of the image (solid lines). (b)Capacitance values of the interface during application of the potential step. Blacktraces: No liposomes in solution. Green and blue traces: with liposomes in solu-tion, separate trials. Letters A-E correspond to time points of images reproducedin Figure 5.4. Blue-shaded background highlights the time period spent at the po-ration potential (-0.4 V/SCE), orange shaded background highlights the time pe-riod at -0.2 V/SCE, and the unshaded background represents time periods spent at0 V/SCE. Capacitance measurements were performed with a 5 mV RMS poten-tial perturbation and a 200 Hz perturbation frequency. Fluorescence images weretaken with either of the exposure settings described in Section 5.1 and converted toequivalent kilocounts per second.1365.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minFigure 5.4: Representative fluorescence images taken during application of a po-tential profile with a 1 min poration step. Images are false colored and contrastenhanced for visibility in print - calibration bars on the right indicate minimumand maximum intensity values presented in each series, in kcts/sec. Scale bars are20 ?m long. Letters A-E correspond to time points marked in Figure 5.3: A - ini-tial image at 0 V/SCE, B - end of 0 V/SCE step, C - end of -0.4 V/SCE step, D- end of -0.2 V/SCE step, E - final image (0 V/SCE). Top row: with liposomes insolution (corresponds to green trace in Figure 5.3). Middle row: With liposomes insolution (corresponds to blue trace in Figure 5.3). Bottom row: without liposomesin solution (corresponds to black trace in Figure 5.3). White outlines in column Aoutline the image areas identified as hotspots for each experiment.1375.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minsurface, and processes such as dimer formation which will reduce the number ofmolecules fluorescing in the spectral range selected.After exposure to liposomes and the poration procedure, the desorption of theadsorbed layer shows significant differences in the potential dependence of fluo-rescence and its increase in intensity. In both examples (blue and green lines inFigure 5.5), the initial fluorescence values (at +0.15 V/SCE) are higher than in thecontrol layer. Partially, this is caused by selection of a naturally more fluorescentregion of the electrode, as identified in the initial image intensities. This effect iscompounded by the increase in fluorescence due to liposome incorporation suchthat both liposome-exposed layers had a substantially increased fluorescence valueat the beginning of the desorption scan. For both layers, the onset of desorption oc-curs at -0.45 V/SCE, coincident with the increase in fluorescence. The shift in theonset of the capacitance and fluorescence increase suggests that the modified layerhas interacted with the liposomes, changing the layer organization and enablingdesorption at less negative potentials than in the unmodified layer. The increasedfluorescence intensity at desorption is indicative of a modified layer structure wheremore of the fluorophore is further from the electrode surface, reducing quenchingand increasing the signal. These fluorescence images also show a less uniformstructure, with the brighter regions highlighting the regions furthest from the elec-trode surface.The layer is desorbed at the negative scan limit, with fluorescence significantlyincreasing and a maximum at -0.65 V/SCE for one layer (blue lines in Figure 5.5),or a steady increase until -0.8 V/SCE in the other (green lines in Figure 5.5). Theappearance of a fluorescence maximum has been observed previously [184], and isbelieved to result from a change in fluorophore organization such as dimer forma-tion. Both layers have a much greater relative fluorescence increase at desorptionthan the control layer, which can be explained by variations in the amount of flu-orophore present in the layer at deposition, combined with a significant change inthe structure of the layer after exposure to the liposomes and poration potentialprofile. The layer has apparently been modified by inclusion of liposomes to createstructures extending normal to the electrode surface (i.e. 3-D structures such as thehemiliposome structure proposed in Figure 5.1).The hotspots in the liposome-exposed layers do not show a significantly differ-ent behaviour than the remainder of the layer, aside from a slightly higher fluores-cence value. Thus, although these hotspots might have been expected to be defec-1385.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 mintive regions in the layer that would be easily accessible to liposome incorporation,more incorporation was not observed in these regions. Electrochemically formeddefects throughout the non-hotspot regions of the layer are therefore believed to bethe main sites for liposome incorporation.Examination of the structures visible in the fluorescence images (Figure 5.6)provides further details on the mechanism of the liposome interaction with the ad-sorbed octadecanol layer. Although the hotspots identified from the initially de-posited layer remain visible throughout the desorption process, during desorption,the remainder of the visible layer undergoes a significant increase in fluorescenceintensity as well. These changes do not appear to be uniform across the image area,however, and appear to consist of small highly fluorescent regions distributed onthe surface. A further study of these features is important as the liposomes may notincorporate evenly across the electrode surface, and their influence on the modifiedlayer structure may be localized. The general increase in fluorescence during des-orption masks these smaller details and makes it difficult to separate the creationof heterogeneity in the fluorescence images from the overall desorption process.Analysis of these small features can be accomplished by removing the featurelessbackground intensity through a rolling ball background subtraction (with a 50 pixel(14 ?m) radius), making the small features easier to distinguish. This backgroundsubtraction was performed on the images of Figure 5.6, and the resulting imagesare presented in Figure 5.7.For the layer that was not exposed to liposomes, the number of bright fea-tures at desorption is essentially the same as at adsorption (features outlined inwhite). Only a few relatively low intensity features are generated at -0.8 V/SCE,and are attributed to structures present in the layer at adsorption that were initiallytoo dimly fluorescent to see due to quenching in the adsorbed layer. In contrast,many new bright features appear even in the images just at the onset of desorptionat -0.45 V/SCE for the layers exposed to liposomes. These structures, not present inthe initially adsorbed layer, are a result of interaction with the liposomes. In bothof the examples that were exposed to liposomes and the poration procedure, thenumber of small fluorescent regions also chagned with the applied potential. As re-gions that are further from the electrode will increase more rapidly in fluorescenceas a result of the nonlinear relationship between quenching and fluorophore-metalseparation (Section 1.6.1), seeing a substantial increase in fluorescence for these re-gions. These small features (~ 5 pixels in diameter - ~2 ?m) are much larger in size1395.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minFigure 5.5: Spectroelectrochemical behaviour of adsorbed octadecanol bilayerswith and without liposomes in solution during desorption of the layer after mod-ification by a 1 min poration potential profile. Potential was scanned from +0.15to -0.8 V/SCE. (a) Fluorescence intensity as average grayscale values for hotspotregions (dotted lines) and the remainder of the image (solid lines). (b) Capacitancevalues of the interface during the potential scan. Black traces: No liposomes insolution. Green and blue traces: with liposomes in solution, separate trials. Ca-pacitance measurements were performed with a 5 mV RMS potential perturbation,a 200 Hz perturbation frequency, and a 5 mV/s potential scan rate. Fluorescenceimages were taken with a 2.5 second exposure time and electron-multiplier gain of200, and converted to equivalent kilocounts per second.than an individual liposome, and so cannot be assigned to one specific interactionevent. Importantly, most of these features are not present in the initially adsorbedlayer and the control layers not exposed to liposomes do not show similar brightregions when treated with the same analysis. Moreover, a distinct difference in the1405.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minFigure 5.6: Representative fluorescence images taken during desorption of the oc-tadecanol layer after modification by a 1 min poration potential profile. Imagesare false colored and contrast enhanced for visibility in print - calibration bars onthe right indicate minimum and maximum intensity values presented in each seriesin kcts/sec. Scale bars are 20 ?m. Top row: with liposomes in solution (cor-responds to green trace in Figure 5.3). Middle row: With liposomes in solution(corresponds to blue trace in Figure 5.3). Bottom row: without liposomes in so-lution (corresponds to black trace in Figure 5.3). White outlines highlight hotspotregions identified in the initial fluorescence images.1415.3. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 1 minFigure 5.7: Fluorescence images of Figure 5.6 after background subtraction by a50 pixel radius rolling ball filter. White outlines highlight hotspot regions identi-fied in the initial fluorescence images. Top row: with liposomes in solution (cor-responds to green trace in Figure 5.3). Middle row: With liposomes in solution(corresponds to blue trace in Figure 5.3). Bottom row: without liposomes in solu-tion (corresponds to black trace in Figure 5.3). Scale bars are 20 ?m.1425.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minnumber of these features is evident when comparing the two layers which were ex-posed to liposomes, indicating that the potential controlled liposome interaction isdependent on the structure of the adsorbed layer, which is not uniform across theelectrode surface.5.4 In-Situ Fluorescence Imaging: Poration at-0.4 V/SCE for 15 minThe changes observed in the adsorbed layer due to the interaction of liposomesshould depend on the time spent at the poration potential (-0.4 V/SCE), resultingin an increased possibility of liposome interaction with increasing poration time.From the electrochemical measurements, increasing the time from 1 min to 15 minresulted in adsorbed layers that were still intact, with capacitance values that didnot change significantly after the poration process unless liposomes were presentin the sub-phase. Also, desorption of the adsorbed layer showed a shift in the po-tential of the onset of desorption that suggested a change in the adsorbed layer afterinteraction with liposomes. The difference between the 1 and 15 minute porationtimes was small, but the increased liposome concentration in the microscopy cell isexpected to increase the chance of interaction with the adsorbed layer, enhancingany differences in the poration procedure.5.4.1 Fluorescence During Potential StepsA potential profile identical to those used in Section 4.4 was used to expose an ad-sorbed octadecanol bilayer to a poration potential of -0.4 V/SCE for 15 min, withand without liposomes in solution. The capacitance and average fluorescence inten-sity values are presented in Figure 5.8. As in the previous data sets, any ?hotspots?or intensely fluorescent areas visible in the initially formed layer are outlined andanlayzed separately.The capacitance changes due to poration are larger than observed in the electro-chemical measurements, but the trends are similar. In the control experiment withno liposomes present (black lines in Figure 5.8), the adsorbed layer capacitanceincreases when the potential is changed to -0.4 V/SCE, and a sharp increase in ca-pacitance is also observed at about 7 min after the move to -0.4 V/SCE. This can beexplained as a change in the layer, but since no change in the fluorescence intensity1435.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minFigure 5.8: Spectroelectrochemical behaviour of adsorbed octadecanol bilayerswith and without liposomes in solution during application of a potential profilewith a 15 min poration step. (a) Fluorescence intensity as average grayscale valuesfor hotspot regions (dotted lines) and the remainder of the image (solid lines). (b)Differential capacitance values of the interface during application of the potentialstep. Black traces: No liposomes in solution. Green traces: with liposomes in so-lution. Letters A-E correspond to time points of images reproduced in Figure 5.9.Blue-shaded background highlights the time period spent at the poration potential(-0.4 V/SCE), orange shaded background highlights the time period at -0.2 V/SCE,and the unshaded background represents time periods spent at 0 V/SCE. Capaci-tance measurements were performed with a 5 mV RMS potential perturbation and a200 Hz perturbation frequency. Fluorescence images were taken with the exposuresettings described in Section 5.1 and converted to equivalent kilocounts per second.was seen (for the visible region), it is more likely due to a change in the wetting ofthe sides of the electrode held in a hanging meniscus. The capacitance decreases1445.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minFigure 5.9: Representative fluorescence images taken during application of a po-tential profile with a 15 min poration step. Images are false colored and contrastenhanced for visibility in print - calibration bars on the right indicate minimum andmaximum intensity values presented in each series in kcts/sec. Letters A-E corre-spond to time points marked in Figure 5.3: A - initial image at 0 V/SCE, B - endof 0 V/SCE step, C - end of -0.4 V/SCE step, D - end of -0.2 V/SCE step, E - finalimage (0 V/SCE). Top row: with liposomes in solution (corresponds to green tracein Figure 5.3). Middle row: With liposomes in solution (corresponds to blue tracein Figure 5.3). Bottom row: without liposomes in solution (corresponds to blacktrace in Figure 5.3). White outlines in column A indicate the image areas identifiedas hotspots for each experiment. Scale bars are 20 ?m.1455.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minduring the time at -0.2 V/SCE and when returning to 0 V/SCE, to value that isslightly above its starting value. In the absence of liposomes in the electrolyte, theadsorbed layer did not experience significant changes during this extended porationprocess.The fluorescence intensity initially decreases due to photobleaching, with thehotspots bleaching more quickly, as observed previously. As in the 1 min controlin Figure 5.3, a slight increase in fluorescence is associated with the change inpotential to -0.4 V/SCE, as expected because the adsorbed layer will slightly changeits organization at these potentials. The origin of these small changes could be adisplacement of fluorophore to the outer surface of the adsorbed layer.In the presence of liposomes (green lines in Figure 5.8), the adsorbed layershowed more distinctive changes, similar to those exposed to the poration poten-tial for 1 min. On changing the potential to -0.4 V/SCE, the capacitance increasesdramatically and significant jumps in capacitance are observed during the 15 minat the poration potential. Similar events were observed for organic droplet adsorp-tion and bursting onto a bare Hg drop[136], but cannot be used to explain thesechanges since the electrode surface is coated in an organic layer. The capacitancedecreases when the potential is changed to -0.2 V/SCE and to 0 V/SCE, resultingin an adsorbed layer that has a capacitance of 5 ?F/cm2, a larger change than in theelectrochemical measurements, although this is most likely due to the increase inliposome concentration when moving to the microscope cell.When compared to the 1 min poration time described above, the changes influorescence for the 15 min poration time are much greater. The experiment shownin Figure 5.8 had very few identifiable hotspots in the initial image, and their be-haviour was identical to the overall fluorescence response, so the profile for thehotspots is not shown here. The initial 15 min spent at 0 V/SCE showed typicalphotobleaching effects. On changing the potential to -0.4 V/SCE, the fluorescenceintensity began a steady increase, distributed evenly across the visible image area.The fluorescence continues to increase steadily with time, showing small decreasesas the potential is changed to -0.2 V/SCE and again to 0 V/SCE, again mostlyevenly across the image area although there is a slightly stronger increase on theright side of the image, visible in columns D and E in Figure 5.9. The steady fluo-rescence increase may be due to diffusion of fluorophore into liposome structuresafter interaction with the adsorbed layer. These structures are expected to extendfarther from the electrode surface than the adsorbed layer, so as fluorophore dif-1465.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minfuses into them, the fluorescence quenching will be reduced, making the intensitybrighter. A similar increase was seen in the 1 min poration experiments in Section5.3, although the effect was not as large as that seen here.5.4.2 Fluorescence During Desorption of the Modified LayersAfter exposure to the poration potential profile, the modified octadecanol bilayerwas desorbed by applying a potential scan from +0.15 to -0.8 V/SCE while takingcapacitance and fluorescence measurements. The average fluorescence and capaci-tance values are plotted in Figure 5.10. The behaviour of the control layer is similarto that of the 1 min exposure, with an increase in fluorescence intensity associatedwith the increase in capacitance accompanying desorption of the layer. Similarly,this indicates that the poration process did not significantly perturb the adsorbedlayer in the absence of liposomes, even though the capacitance increased slightlyin the poration process.For the example exposed to liposomes (green lines in Figure 5.10), the desorp-tion behaviour was very different. The capacitance shows the onset of desorptionat -0.45 V/SCE, with full desorption attained by -0.8 V/SCE, as signified by thecapacitance reaching a value similar to that for a water covered electrode. Thefluorescence increased continuously prior to desorption, with a rapid increase be-ginning around -0.5 V/SCE. The change in fluorescence is most sensitive to thefeatures that are furthest from the electrode surface since these features producethe largest fluorescence signals - small changes in the distance from the electroderesult in large changes in the fluorescence. This indicates that features of the modi-fied layer are indeed farther from the electrode surface. At potentials more positivethan -0.5 V/SCE, although the fluorescence is increasing, the capacitance changesonly slightly - capacitance is most sensitive to changes in the dielectric of mate-rial on the electrode surface and less sensitive to changes farther from the surface.At potentials more negative that -0.5 V/SCE, both the capacitance and fluorescenceincrease sharply as the layer becomes desorbed. Although the details of the fluores-cence behaviour are specific to the area being imaged, making direct comparisonsdifficult, the similarity of the general behaviours between the 1 min and 15 minporation times are clear.As in the 1 min poration examples, for both the layers with and without lipo-somes in solution the increase in fluorescence is fairly uniform across the imaging1475.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minFigure 5.10: Spectroelectrochemical behaviour of adsorbed octadecanol bilay-ers with and without liposomes in solution during desorption of the layer aftermodification by a 15 min poration potential profile. Potential was scanned from+0.15 V/SCE to -0.8 V/SCE. (a) Fluorescence intensity as average grayscale val-ues for hotspot regions (dotted lines) and the remainder of the image (solid lines).(b) Differential capacitance values of the interface during the potential scan. Blacklines: No liposomes in solution. Green lines: with liposomes in solution. Capac-itance measurements were performed with a 5 mV RMS potential perturbation at200 Hz and a 5 mV/s potential scan rate. Fluorescence images were taken witha 2.5 second exposure time and electron-multiplier gain of 200, and converted toequivalent kilocounts per second.area (Figure 5.11), although the appearance of some regions of more intense fluo-rescence are visible. Removal of the general fluorescence increase by applying arolling ball filter as described previously helps highlight the changes in structurein the images as the layer is desorbed. A comparison of the changes in the small1485.4. In-Situ Fluorescence Imaging: Poration at -0.4 V/SCE for 15 minFigure 5.11: Representative fluorescence images taken during desorption of theoctadecanol layer, after modification by a 15 min poration potential profile. Imagesare false colored and contrast enhanced for visibility in print - calibration bars onthe right indicate minimum and maximum intensity values presented in each seriesin kcts/sec. Top row: with liposomes in solution (corresponds to green trace inFigure 5.10). Bottom row: without liposomes in solution (corresponds to blacktrace in Figure 5.10). White outlines in leftmost column highlight hotspot regionsas identified in the initial fluorescence images. Scale bars are 20 ?m.Figure 5.12: Fluorescence images of Figure 5.11 after background subtraction bya 50 pixel radius rolling ball filter. White outlines highlight hotspot regions asidentified in the initial fluorescence images. Top row: with liposomes in solution(corresponds to green trace in Figure 5.10). Bottom row: without liposomes insolution (corresponds to black trace in Figure 5.10). Scale bars are 20 ?m. Hotspotregions are highlighted by white outlines in all images.1495.5. Conclusionsfeatures observed both with and without liposomes in solution is shown in Figure5.12. The outlined features are the ?hotspots? identified at the start of the in the ini-tially adsorbed layer before poration. The control layer without liposomes showsvery few increases in the number or size of these features after poration, annealingand desorption. In contrast, the layer that was exposed to liposomes through theporation procedure displays a significant number of these features, far above thatobserved before the poration process. Even while adsorbed at 0 V/SCE after pora-tion, but before desorption, (rightmost column in Figure 5.12), the appearance ofbrightly fluorescent regions is clear. These features are similar to those observed inthe 1 min poration studies, with a density that is between the two 1 min porationexamples. Even though it was expected that the time allowed for liposome incorpo-ration should increase the number of features observed as well as the fluorescenceintensity, the small region chosen for analysis has a large influence on the extentof liposome interaction and incorporation observed, making a comparative analysisdifficult.5.5 ConclusionsFluorescence microscopy combined in-situ with electrochemical measurement wasused to observe the interaction of liposomes with an adsorbed octadecanol bilayer.The fluorescence behaviour confirms the observations of the electrochemical mea-surements in Chapter 4 that liposome interaction with the adsorbed layer dependson the potential controlled creation of defects in the adsorbed layer.In the cases where liposomes are present in the electrolyte, the small increasesin the adsorbed layer capacitance after the poration procedure coincides with in-creases in fluorescence due to a redistribution of fluorophore after liposome in-teraction. The incorporation of liposomes into the adsorbed layer was found tochange the desorption potential of the adsorbed layer, desorbing at less negativepotentials due to the defects created through liposome interaction. Fluorescenceimaging of the desorption process reveals the presence of small structures that maybe regions where liposomes are incorporated. These structures are strongly in-fluenced by the initial quality or nature of the layer deposited, which was foundto be neither uniform nor homogeneous. Although the structures observed byfluorescence are larger than an individual liposome, investigation using atomicforce microscopy (AFM) may reveal further information on the exact nature of1505.5. Conclusionsthe liposome-octadecanol interaction.151Chapter 6Atomic Force MicroscopyInvestigations ofOctadecanol-coated SurfacesAs demonstrated in Chapter 5, fluorescence microscopy is a powerful tool for ob-serving the electrode-solution interface. However, it is limited in resolution bydiffraction, and without detailed quenching curves cannot provide exact informa-tion on fluorophore-electrode separation. Atomic force microscopy (AFM) pro-vides a much higher lateral resolution - much more than needed to resolve 100 nmliposomes. Fluorescence is only able to provide limited information on structuresraised above the electrode surface, however by its nature, AFM can provide directinformation on the three dimensional size and shape of structures on the surface.When using intermittent contact mode, analysis of the phase of the cantilever oscil-lation can also provide some information on surface characteristics via tip-surfaceinteraction.To this end, exploratory measurements were done using AFM in ex-situ (air)and in-situ (in liquid, under potential control) on octadecanol layers adsorbed ontogold substrates. These measurements are complimentary to those done with fluo-rescence.6.1 Ex-situ Imaging of Octadecanol Monolayers on AuOctadecanol monolayers have been shown, both in the literature [40, 162] and in thework presented in this thesis, to be non-homogenous in nature despite their repro-ducible electrochemical properties. Before proceeding to more complex measure-ments, the typical characteristics of octadecanol adsorbed on an electrode surfacein AFM must be established.1526.2. Correlation with Fluorescence Imaging6.1.1 Experimental MethodsMonolayers of octadecanol containing 3 mol% of the BODIPY-C19-OH fluorophorewere prepared on either a Au bead with Au(111) facet or Au/mica substrate as de-scribed in Section 2.7. The prepared substrates were then allowed to dry in air be-fore imaging by AFM. All images were performed in intermittent contact mode, us-ing either ACAFM with VistaProbes cantilevers (model T190R-25, nominal forceconstant 48 N/m, resonant frequency 190 kHz) or MAC mode with Type II can-tilevers (Agilent, nominal force constant 2.8 N/m, resonant frequency 75 kHz). Allmeasurements were performed in air under ambient conditions.Topography images were processed using Gwyddion [214]. First a median linecorrection was applied, followed by a polynomial background subtraction (degreeof 2 in both x and y). The lowest measurement value was set to 0, and the imagecontrast adjusted. Phase images were treated only with the median line flatten andcontrast adjustment.6.1.2 ResultsThe adsorbed monolayers of octadecanol including 3 mol% BODIPY-C19-OH flu-orophore showed a small variety of structures upon deposition. Two representativeimages, one from each substrate, are shown in Figures 6.1 and 6.2. In some regions,fine structural details of the octadecanol layer structure could be observed, such asthe small holes or indentations visible in Figure 6.1. A common motif was rela-tively large ?blobs? scattered on the surface, as visible in Figure 6.2, and to a lesserdegree in Figure 6.1. These blobs are visible as raised regions in the topography,and also show in the phase imaging as distinct regions, suggesting that they have adifferent tip-sample interaction compared to the rest of the layer. These blobs aremost likely small regions of multilayer octadecanol, which might be softer or morecompressible than the surrounding monolayer regions, resulting in the contrast ob-served in the phase measurement.6.2 Correlation with Fluorescence ImagingAlthough measurements correlating fluorescence and AFM imagery have been pre-viously shown [215, 216], the exact nature of the structures seen in the adsorbedoctadecanol layers is unknown. The fluorescent ?hotspots? noted in previous chap-1536.2. Correlation with Fluorescence Imaging(a) Topography (b) PhaseFigure 6.1: AFM image of a n octadecanol monolayer containing 3 mol%BODIPY-C19-OH on the Au(111) facet of a bead electrode. Images were acquiredin ACAFM mode in air.(a) Topography (b) PhaseFigure 6.2: AFM image of an octadecanol monolayer containing 3 mol% BODIPY-C19-OH on a Au/mica substrate. Image acquired in MAC mode in air.1546.2. Correlation with Fluorescence Imagingters are presumed to be due to 3-D structures separating fluorophore from the elec-trode surface, but could also be caused by 2-D aggregation of fluorophores withinthe layer locally increasing the fluorophore concentration. A correlation of the flu-orescence images to the AFM should give some idea of the nature of the fluorescentstructures seen in an adsorbed octadecanol layer.6.2.1 Experimental MethodsA monolayer of octadecanol containing 3 mol% BODIPY-C19-OH was depositedonto the Au(111) facet of the bead substrate, as described in Section 2.7. Aftermounting in the sample holder, the bead was allowed to dry for approximately onehour in air. An area of the facet surface was imaged by AFM, using ACAFM mode(intermittent contact mode) and VistaProbes tips (model T190R-25, nominal forceconstant 48 N/m, resonant frequency 190 kHz). The approximate cantilever loca-tion and orientation on the surface was recorded using the AFM-mounted camera.The sample was then removed from the AFM and transferred to the fluorescencemicroscope. Fluorescence and brightfield images of the surface were acquired inthe region near the AFM imaging location. Both brightfield and fluorescence im-ages were taken using the 50x objective.Topography images were processed using Gwyddion [214]. First a median linecorrection was applied, followed by a polynomial background subtraction (degreeof 2 in both x and y) and a ?remove scars? filter. The lowest measurement pointwas set to 0, and the image contrast adjusted.Alignment and analysis of the images was performed in GIMP [217]. Thefluorescence, brightfield, AFM camera, and topography AFM images were openedas stacked layers in a single image. The fluorescence and brightfield data, as wellas that from the AFM camera, were scaled to have the same pixel scale as the AFMimage. The AFM camera image was rotated so that visible surface features alignedwith the brightfield image, and the topography channel of the AFM topographyimage rotated to match. The position and rotation angle of the topography imagewas manually adjusted to align visible features in the topography to similar featuresin the fluorescence image. A mask of features visible in the topography image wasmade by hand-tracing the features onto another image layer using the pencil tool.1556.2. Correlation with Fluorescence ImagingFigure 6.3: AFM topography image (left) of the Au(111) facet with octadecanolmonolayer. (Right) Mask of major topographical features from the AFM image.6.2.2 ResultsEx-situ fluorescence and AFM measurements were performed on a monolayer ofoctadecanol containing 3 mol% of the BODIPY-C19-OH fluorophore. The AFMtopography image acquired is shown in Figure 6.3.The background texture of theoctadecanol layer on the Au-mica terraces, similar to that seen in Figures 6.1 and6.2, is visible. Some more distinct features are also visible, as outlined in themask on the left in Figure 6.3. Most of these features were well-defined regions,however a few, such as the diagonal stripe in the top third of the image, and thelarge blob in the top right, appear to be composed at least in part of loose materialthat moved with the tip as it scanned the surface. Two notable features, the largestripe originating at the bottom of the image and the v-shape against the right imageborder, were actually flatter regions that lacked the typical ?octadecanol texture?found in the rest of the image, rather than raised regions.Using the methods described above, these features identified from the AFMimage were overlaid with the same region in the fluorescence image. The fluo-rescence image and overlay of the AFM features onto it are shown in Figure 6.4.Several of the topographically visible features, such as the large hotspot near thecentre of the fluorescence image, as well as the dark streaks at the right hand side,correlate across the two images. Features from the AFM image mask that matchwith fluorescent features are outlined in green in the overlay in Figure 6.4. How-ever, there are a number of features visible to AFM that do not appear to correspondwith fluorescent features, and vice versa.1566.3. In-situ Imaging under Potential ControlFigure 6.4: Fluorescence image of the region surrounding the area imaged in Figure6.3. Left: Fluorescence image (false color). Scale bar is 20 ?m. Right: Fluores-cence image overlaid with the mask of AFM features. Mask areas that overlap afluorescence feature are outlined in green.The results of this correlation show that some fluorescence features are visiblein an AFM topography image, as expected for hotspots that are caused by 3-D ag-gregates of fluorescent material on the surface. Some lower-fluorescence regionsare visible as depressed regions in the AFM image, such as the curved stripes,which may be caused by cracks in the octadecanol layer formed at deposition. Anumber of features are visible to topography but not fluorescence, as would be ex-pected for aggregates of non-fluorescent material on the surface. As well, manyfluorescence features were not visible on the topography image. This may indicatethat they are not 3-D aggregates as hypothesized, but possibly regions of higherfluorophore concentration within the monolayer. However it is also possible thatthese regions are raised away from the electrode surface, but the octadecanol layeris soft enough that it was deformed during imaging, making regions with a smallheight difference difficult to discern. Imaging of the octadecanol under liquid witha cantilever of lower force constant would be required in order to remove this am-biguity.6.3 In-situ Imaging under Potential ControlWith the characteristics of octadecanol layers established, the ground is laid forin-situ imaging under potential control. Roughly replicating the experiments per-1576.3. In-situ Imaging under Potential Controlformed with fluorescence microscopy, these experiments provide a higher-resolutionview of the interaction of octadecanol with the electrode surface and with lipo-somes.6.3.1 Experimental MethodsAll in-situ imaging was done using Au/mica substrates, prepared with a bilayerof octadecanol containing 3 mol% BODIPY-C19-OH as described in Section 2.7.The substrate was then mounted in the liquid cell, which was filled with 0.1 M NaFelectrolyte. If required, 1 ?L of the liposome solution was then added. The entiresample assembly was placed in the environmental chamber and allowed to rest in anitrogen atmosphere for one hour to minimize oxygen in solution before imaging.Potential was held constant during imaging. After completing two images, the po-tential was stepped directly to the next value, and the system allowed to equilibratefor approximately 10 seconds before imaging again. All potentials were measuredagainst an Au bead pseudoreference. Based on the position of Au oxidation peaks(measured independently), this reference electrode is approximately -0.1 V/SCE.Therefore the 0 and -0.5 V/Au potentials used are approximately equal to 0.1 and-0.4 V/SCE, similar to what was used in the electrochemical and fluorescence stud-ies.Although the substrate was not moved during the measurements, accidentalcontact between the cantilever holder assembly and the reference or counter elec-trodes required the tip to be lifted and re-engaged with the surface. During theoctadecanol-only experiment, this resulted in slightly different areas being imagedat each potential. This drift was less of a problem during the measurements withliposomes in solution due to a better initial sample placement, so the area imagedat each potential is nearly identical.All in-situ images were taken in intermittent contact MAC mode. The exper-iment without liposomes in solution was imaged using Type II MAC cantilevers(Agilent, nominal force constant 2.8 N/m, resonant frequency 75 kHz in air). Theexperiment with liposomes in solution used a Type I MAC cantilever (Agilent,nominal force constant 0.6 N/m, resonant frequency 75 kHz in air).AFM images were processed using Gwyddion. The images were subjectedto a median line correction followed by a second-degree polynomial backgroundsubtraction. The minimum height value was set to zero and contrast adjusted for1586.3. In-situ Imaging under Potential Controlvisibility. The images with liposomes in solution were originally taken as 3 ?msquare images, and cropped to 2 ?m squares for a more direct comparison withthe liposome-free images. Particle heights were obtained by manually drawing lineprofiles through the center of the feature in question, and the height taken as theaverage of the height measured at each edge of the feature.6.3.2 ResultsThe topography images acquired for an octadecanol bilayer with no liposomes insolution are presented in Figure 6.5. A sample of representative height profiles andaverage values for different features seen in the images is shown in Figure 6.7.Little change in the layer is seen as the potential is stepped from 0 to -0.5 V/Auand back. Initially (Figure 6.5a) the same low flat features as previously seen in theex-situ octadecanol layers are visible. At -0.5 V/Au, similar features are seen, atapproximately the same height, as seen in the profiles of Figure 6.7. Although atthis potential defects should be present in the octadecanol layer, none are visible inthe images. This may be because the defects are too small in either height or diam-eter to be detected with the AFM, or because the force of the cantilever contactingthe surface is enough to blur out the defect edges and render them invisible.Upon return to 0 V/Au, where the potential-created defects will be healed, afew of the low flat blobs visible in other images are still present at a reduced height,but the layer is mostly flat and featureless. Generally, all of the visible features areslightly smaller than their counterparts at the initial -0.5 V/Au images (See Figure6.7). This is not necessarily indicative of a change in the layer itself, as the imagingarea is different from the previous two images. The layer may have flattened afterapplication of the potential profile, or it may simply be a more featureless region ofthe layer being imaged.Analysis of the phase channel of the images (Figure 6.6) shows that, as demon-strated in the ex-situ measurements, the low blobs exhibit a visibly different phasebehaviour than the surrounding layer..A repeat of the same potential profile with liposomes in solution was also per-formed. In this case, the imaging area remained the same across all potentials, sofeatures can be tracked more directly. The topographical images are shown in Fig-ure 6.8, and sample cross sectional profiles of the features are shown in Figures 6.101596.3.In-situImagingunderPotentialControl(a) 0 V / Au (initial). (b) -0.5 V/Au. (c) 0 V/Au (final).Figure 6.5: In-situ AFM images of an octadecanol bilayer containing 3 mol% BODIPY-C19-OH: topography channel. Potentials aremeasured vs. an Au bead reference.1606.3.In-situImagingunderPotentialControl(a) 0 V/Au (initial) (b) -0.5 V/Au (c) 0 V/Au (final)Figure 6.6: In-situ AFM images of an octadecanol bilayer containing 3 mol% BODIPY-C19-OH: phase channel. Potentials aremeasured vs. an Au bead reference.1616.3.In-situImagingunderPotentialControlFigure 6.7: Example features of each category from the AFM topography images of octadecanol bilayers. Images are cropped fromthose in Figure 6.5. Profiles shown in plots correspond to the cross-sections drawn on each image. Features in each category weremeasured on the full images to provide average height values at each potential. Scale bars are 200 nm.1626.3.In-situImagingunderPotentialControl(a) 0 V/Au (initial). (b) -0.5 V/Au. (c) 0 V/Au (final).Figure 6.8: In-situ AFM images of an octadecanol bilayer containing 3 mol% BODIPY-C19-OH and liposomes in solution: topog-raphy channel. Potentials are measured vs. an Au bead reference.1636.3.In-situImagingunderPotentialControl(a) 0 V/Au (initial) (b) -0.5 V/Au (c) 0 V/Au (final)Figure 6.9: In-situ AFM images of an octadecanol bilayer containing 3 mol% BODIPY-C19-OH and liposomes in solution: phasechannel. Potentials are measured vs. an Au bead reference.1646.3.In-situImagingunderPotentialControlFigure 6.10: Example features that were present in the initially deposited layer of each category from the AFM topography images ofoctadecanol bilayers with liposomes in solution. Images are cropped from those in Figure 6.8. Profiles shown in plots correspond tothe cross-sections drawn on each image. The features measured in the 0 V (initial) images are re-measured at the other two potentials.Features in each category were measured on the full images to provide average height values at each potential. The ?small features?were not visible except in the initial 0 V image. Scale bars are 200 nm.1656.3.In-situImagingunderPotentialControlFigure 6.11: Example features that appeared during liposome adsorption from the AFM topography images of octadecanol bilayers.Images are cropped from those in Figure 6.8. Profiles shown in plots correspond to the cross-sections drawn on each image. Thefeatures measured in the 0 V (initial) images are re-measured at the other two potentials (image #1 for each). Image #2 in the lasttwo potentials is a feature that was not visible at the initial 0 V image - the same feature in both -0.5 V and 0 V (final). Features ineach category were measured on the full images to provide average height values at each potential. The ?small features? were notvisible except in the initial 0 V image. Scale bars are 200 nm.1666.3. In-situ Imaging under Potential Controland 6.11. Since it was possible to image the same region at all potentials, an anal-ysis of features visible in all images is presented alongside features that appearedonly after poration in the profiles of Figure 6.10.Initially, the layer is relatively flat, with a few larger features and the flat blobstypical of an octadecanol layer. At -0.5 V/Au, a large number of new featuresappear. Most of these features are significantly taller than the flat blobs previouslyseen, although some features of a similar height also appear. This can be seen inthe profiles in Figures 6.10 and 6.11, where the average heights of both the flat blobfeatures and the tall features is increased over those measured at 0 V/Au. Many ofthese new features persist upon return to 0 V/Au, and retain some of the increasedheight.The appearance of these taller features is absent in the octadecanol-only im-ages, and must be due to the interaction of liposomes with the adsorbed layer.None of the feature heights measured is comparable to the 100 nm size expected ofa whole liposome adsorbed onto the surface, even accommodating for some distor-tion caused by tip effects on a soft structure. Therefore the features must be regionswhere the liposomes have burst to incorporate into the octadecanol layer, forming3-D structures above the electrode surface. Based on height measurements, thesestructures may be hemiliposomes or some partially flattened structure.Analysis of the phase channel of the AFM images (Figure 6.9) shows anotherremarkable change with liposome incorporation. Whereas in the absence of lipo-somes, the structures in the adsorbed layer are distinguishable in the phase images,upon liposome incorporation the contrast between the topographical features andlayer background is largely lost. This certainly indicates that the structures seen inFigure 6.8b and c are of a different nature than those seen in the unmodified oc-tadecanol layers. It may also be possible that the phase contrast is erased becauseliposomes are depositing across the entire electrode surface on top of the octade-canol layer, causing the entire surface to have similar tip-sample interactions. Thiscannot be determined from the measurements done here, however film thicknessmeasurements, easily performed in-situ, may provide future insight.Comparison with the fluorescence results from Section 5.3.1 provides furtherinsight. By using the same rolling-ball subtraction performed on the desorption im-ages, it is clear that during the poration phase of the experiment, no new structuresare formed in the octadecanol layer without liposomes (Figure 6.12, top). Withliposomes in solution, a large number of new features are formed upon application1676.4. ConclusionsFigure 6.12: Fluorescence images taken during the application of potential stepsafter background subtraction by application of a 50 pixel rolling ball filter. Originalfluorescence images are found in Figure 5.4. Top: without liposomes in solution.Bottom: with liposomes in solution. Scale bar is 20 ?m.of the -0.4V/SCE potential, some of which persist as the potential is changed backto 0 V/SCE. This behaviour is exactly analogous to that observed in the AFM mea-surements. Although the scale of the images is different between the two imagingmodes, they share the same potential-dependent changes. If the changes in flu-orescence intensity are proportional to changes in height, then the changes seenin the fluorescence images of Figure 6.12 can be interpreted as showing the samebehaviour seen in the AFM images.6.4 ConclusionsAtomic force microscopy, both in-situ and ex-situ, has been performed on solid-supported octadecanol layers. Ex-situ measurements show that the layer is non-uniform, containing structures of approximately 15 nm height that are visible toboth topography and phase contrast imaging. Some features visible to AFM imag-ing can be identified as fluorescent regions in the layer, however a complete corre-lation is not possible.In-situ imaging of the octadecanol bilayer under potential control showed littlechange in the layer in the absence of liposomes. With liposomes in solution, a num-ber of large features appeared across the image area upon creation of defects in theoctadecanol layer. Many of these structures persisted upon return to 0 V/Au and1686.4. Conclusionsmay be assigned to liposomes incorporating into the defects created in the octade-canol layer. Phase contrast analysis shows no difference between these structuresand the surrounding layer, indicating that the layer after potential perturbation isdifferent with liposomes in solution than in the absence of liposomes.The incorporation of liposomes into the adsorbed octadecanol layer while underpotential control has been directly observed by in-situ atomic force microscopy.The behaviour seen correlates with the previously observed fluorescence behaviour,providing further evidence for the potential-controlled incorporation of liposomesinto the adsorbed octadecanol layer.169Chapter 7Summary and Future WorkThe development of stable, easily modifiable solid-supported layers is critical forthe creation of robust biosensor platforms involving transmembrane proteins andion channels, which are the major targets for modern drug discovery [8]. Includingphospholipid bilayers in the design is essential in order to support these proteins,as their structure and function greatly depends on their environment, and the hy-drophobic regions of the membrane are an important factor. Building the sensordirectly onto an electrode surface builds in the ability to control the transmembranepotential, mimicking the natural cell membrane, however proteins may denatureif in direct contact with an electrode surface. Phospholipid bilayers on their ownare easily and conveniently created, but are not well adhered to electrode surfaces,requiring the use of a supporting molecule or tether to anchor them to the electrodesurface and provide a space for proteins to extend outside the membrane on eitherside, without contacting the metal directly.The work presented in this thesis describes a method for producing supportedphospholipid layers in-situ by control of the electrode potential. This method pro-duces a three-dimensionally structured layer of modified lipid bilayer with someregions separated from the electrode surface. These structures have the potential tohouse transmembrane proteins safely away from the electrode surface without theuse of special tethering molecules, as illustrated by the cartoon in Figure 7.1. Thepresence of a small pocket of electrolyte would make sensors based on this platformsuitable for direct detection of ion channel activity, and as proteins are separatedfrom the electrode surface, fluorescence-based detection of binding events is alsopossible.By using an initially deposited octadecanol bilayer, the electrode surface is pro-tected from liposome deposition until the application of a negative electrode poten-tial. This potential opens defects in the adsorbed octadecanol layer, and allowsthe liposomes to incorporate into the adsorbed layer, forming regions of incorpo-rated liposomal bilayer across the electrode surface. This process was refined and170Chapter 7. Summary and Future WorkFigure 7.1: Cartoon schematic of protein incorporation into a raised liposomal bi-layer structure (not to scale). One possible location for a transmembrane protein isshown in green.characterized as described as follows.The physical, electrochemical and fluorescence properties of adsorbed octade-canol layers modified with a fluorescent probe were characterized in Chapter 3.Few methods allow for gathering information on the properties of an adsorbed sur-face layer in-situ and with spatial resolution. Fluorescence microscopy is one suchtechnique, however it requires that the adsorbed layer under study be either inher-ently fluorescent or doped with a small amount of a fluorescent molecule to makeit visible. Since octadecanol is not fluorescent, a fluorophore must be added, in thiscase the BOCIPY-C19-OH molecule designed specifically for use in octadecanolmonolayers. Compression isotherms of octadecanol monolayers containing vari-ous concentrations of the fluorophore showed that despite its? structural similarityto octadecanol, the BODIPY-C19-OH causes some disruption to the monolayer or-ganization even at low concentrations. At concentrations above 3 mol% however,the disruptions became more pronounced and characteristic of a poorly mixed sys-tem. Electrochemical characterization of monolayers and bilayers of octadecanolwith the fluorophore confirmed this pattern, with 0.5 mol% and 3 mol% show-ing the least altered electrochemical characteristics of the concentrations tested,relative to pure octadecanol. Fluorescence microscopy of the BODIPY-C19-OHcontaining octadecanol bilayers showed that 3 mol% of the fluorophore was suffi-cient to ensure a detectable fluorescence signal, at least at desorption. These resultswere used to design the octadecanol layers used for continuing study of liposomeincorporation.The explorations of the potential-controlled interaction of DOPC liposomeswith the octadecanol bilayer were presented in Chapter 4. Liposomes would notincorporate into the as-deposited octadecanol layer at moderate potentials, evengiven over an hour of exposure. Application of a potential sufficient to cause sig-171Chapter 7. Summary and Future Worknificant disruption to the adsorbed layer structure was required to facilitate interac-tion with the liposomes - in this case a potential more negative than -0.4 V/SCE.The magnitude of the potential used to create defects in the octadecanol layer hadsome effect on the characteristics of the liposome-modified layer, but interestingly,complete desorption of the layer did not increase the apparent degree of liposomeincorporation into the adsorbed layer. As long as a sufficiently large potential wasapplied, the time spent at that potential was found to have only a weak effect onthe interaction of liposomes with the adsorbed layer. Shorter times, such as 1 min,were both convenient and effective at allowing liposomes to incorporate into theoctadecanol layer.Changes in the adsorbed octadecanol layer were easily observed by a shift inthe double-layer capacitance of the electrode as the layer structure was altered dueto changes in surface tension with potential. These measurements are simple andeasy to implement, but do not provide any detail on the exact form of the changesmade to the layer. In-situ fluorescence microscopy, presented in Chapter 5, givesinsight into the changes occurring on the electrode surface. As any fluorescentmaterial near the electrode surface is quenched, changes in layer structure that re-sult in a separation of the adsorbed layer from the surface are highlighted by alocal increase in fluorescence intensity. Observations of the general fluorescenceproperties confirmed that liposomes do not alter the octadecanol layer until afterapplication of the poration potential. Analysis of the fine structure visible in thefluorescence images revealed the creation of small raised (and therefore fluores-cent) structures on the electrode surface whose appearance coincided with the ca-pacitance changes attributed to liposome incorporation. Although the resolution ofoptical microscopy is not sufficient to assign these structures to individual liposomeincorporation events, the formation of these structures is evidence of the incorpo-ration of liposomes at specific locations on the surface, creating a heterogeneouslayer structure.The improved resolution of atomic force microscopy was leveraged in the ex-periments documented in Chapter 6 to provide further detail on the nature of the in-teraction at the electrode surface. The octadecanol layer itself was shown to be quiteheterogeneous, as might be expected from the structures visible to fluorescence mi-croscopy. In-situ imaging provided a unique view into the incorporation process,showing the creation of tall structures during exposure to the poration potential thatremained visible even after a return to 0 V/SCE. Although these structures can not1727.1. Directions for Future Studybe confirmed as the same structures seen in the fluorescence microscopy due to thedifference in scale, they bear a strong similarity and certainly are a part of the sameprocess of liposome incorporation.Overall, these results show that it is possible by using a potential-mediateddeposition process to create, in situ, a multi-component supported bilayer suitablefor use as a biosensor platform. Use of electrochemical potential as a mechanismfor preventing or allowing interaction with an adsorbed layer was demonstratedwith electrochemical evidence alongside fluorescence and atomic force microscopyimages illustrating the process of incorporation and confirming in part the proposedmechanism of formation.7.1 Directions for Future StudyAlthough the mechanism of forming the multi-component layers has been provensound, many avenues for continuing study of this system remain. As this method isintended to become a biosensor platform, adding sensor functionality to the systemis an obvious next step. Sensors based on transmembrane proteins have alreadybeen demonstrated using other models of solid-supported bilayers, by incorporat-ing ion channels both natural [13?15, 218] and synthetic, [16, 219] and transportproteins, [12] among others. These transmembrane proteins are relatively easilyincorporated into liposomes, and could therefore be applied to the modified octade-canol model. By using liposomes prepared with incorporated gated ion channels,for example, a structure similar to that envisioned in Figure 7.1 could be created asan analog to similar sensors made with tethered bilayers. This would allow a di-rect comparison of ease of formation and sensitivity of sensors made with the twoplatforms.The ability to control deposition could be leveraged to deposit different lipo-somes onto the various microelectrodes of an interdigitated electrode chip. Theability to create a single chip with several different supported bilayer functional-ities would be a great advantage in creating multifunctional sensors, especially ifthe bilayers can be added in-situ. Figure 7.2 illustrates how this process could becarried out inside a microfluidic chamber. Proof of concept of this process could beeasily tested before application to a microfluidic chip by using a series of liposomesolutions, with each liposome created with different fluorescent tags. The electrodechip, after modification with the octadecanol bilayer, would be immersed in each1737.1. Directions for Future StudyFigure 7.2: A possible procedure for creating multifunctional sensors in-situ in amicrofluidic cell. Liposomes with three different functionalities are deposited, eachcontaining for example a different ion channel or fluorescent probe. Step (1) showsthe electrodes and microfluidic chamber in the initial condition before liposomeincorporation. In Step (2) the poration potential is applied to one electrode only,while a solution of Liposome A flows through the chamber. Steps (3) and (4)repeat this process using different liposome solutions and poration at sequentialelectrodes. In Step (5), all three electrodes have been modified with liposomes, andthe analyte can interact to produce three simultaeous analytical signals.solution sequentially with potential control to allow liposome incorporation to oneelectrode in each solution. The results of the incorporation could then be easilyverified with fluorescence microscopy and the model applied towards building amultichannel sensor.Before the system demonstrated here could be truly called functional as a sen-sor system, the issues surrounding use of octadecanol as the base layer must beaddressed. Although octadecanol has appropriate electrochemical properties, it isvery sensitive to small levels of contamination in the electrolyte, and even as asingle-component floating monolayer forms phase-segregated structures [40] thatreduce the reproducibility of depositions onto the electrode. Use of a more liquidlayer, such as oleyl alcohol (cis-9-octadecen-1-ol) which is liquid at room tem-perature, or a mixed monolayer of a liquid and a solid-phase component, such asoctadecanol/oleyl alcohol, may improve reproducibility of the deposited layers byreducing phase segregation.1747.1. Directions for Future StudyVariations in the liposome composition will also have an effect on the incor-poration behaviour. Model cell membranes designed to support transmembraneproteins are often made of a mix of phospholipids, and may also include choles-terol. The particular mix of lipids used as well as cholesterol content is known toaffect the rigidity of the liposomes formed [198?200], which should also affect theelectrochemical incorporation properties of the liposomes. Liposome size may alsoinfluence their incorporation characteristics, as smaller vesicles are under greaterpressure due to the increased curvature.Continued study using in-situ AFM would further characterize the incorpora-tion behaviour of the liposomes. Use of force-distance spectroscopy would helpestablish the thickness of the adsorbed layer at various structures, as well as somedetails of its nature (for example, bilayer or multilayer). Characterizing both theinhomogeneities of the octadecanol layer and the newly created liposomal struc-tures in this way would further clarify the nature of the liposome-octadecanol in-teraction, for example determining if a mono- or bilayer is deposited on top of theoctadecanol layer, remaining invisible to fluorescence and affecting the phase con-trast in the AFM images. Force curve mapping would also determine the rigidityof the adsorbed liposomal structures, and the nature of their 3-D arrangement.The ability to incorporate liposomes onto an electrode, mediated by potentialcontrol, opens a wide variety of options for future sensor design. The primary ad-vantages are the ability to create layers in-situ, and the ease of creating bilayersthat contain transmembrane proteins or other membrane components directly on aplatform that allows manipulation of the transmembrane potential. After identify-ing a more robust base layer, this method has the potential to create a wide rangeof biosensors applicable to drug discovery and health care applications.175References[1] Musgrove, A.; Bridges, C.; Sammis, G. M.; Bizzotto, D. Langmuir 2013,29, 3347?3360.[2] Overington, J. P.; Al-Lazikani, B.; Hopkins, A. L. Nat Rev Drug Discov2006, 5, 993?996.[3] Tien, H.; Ottova, A. Electrochimica Acta 1998, 43, 3587 ? 3610.[4] Clark, L. C.; Lyons, C. Annals of the New York Academy of Sciences 1962,102, 29?45.[5] Rasooly, A. Methods 2005, 37, 1 ? 3.[6] Dill, K.; Montgomery, D.; Ghindilis, A.; Schwarzkopf, K.; Ragsdale, S.;Oleinikov, A. Biosensors and Bioelectronics 2004, 20, 736 ? 742.[7] Cheley, S.; Gu, L.-Q.; Bayley, H. Chemistry & Biology 2002, 9, 829 ? 838.[8] Jackman, J. A.; Knoll, W.; Cho, N.-J. Materials 2012, 5, 2637?2657.[9] Sinner, E.-K.; Knoll, W. Current Opinion in Chemical Biology 2001, 5, 705? 711.[10] Tvarozek, V.; Tien, H. T.; Novotny, I.; Hianik, T.; Dlugopolsky, J.;Ziegler, W.; Leitmannova-Ottova, A.; Jakabovic, J.; Rehacek, V.; Uhlar, M.Sensors and Actuators B: Chemical 1994, 19, 597 ? 602.[11] He, Y.-E.; Xie, M.-G.; Ottova, A.; Ti Tien, H. Analytical Letters 1995, 28,443?450.[12] Seelheim, P.; Galla, H.-J. Biochemical and Biophysical Research Communi-cations 2013, 431, 519 ? 523.176References[13] Stenger, D. A.; Fare, T. L.; Cribbs, D. H.; Rusin, K. M. Biosensors andBioelectronics 1992, 7, 11 ? 20.[14] Stora, T.; Lakey, J. H.; Vogel, H. Angewandte Chemie International Edition1999, 38, 389?392.[15] Zhang, Y. L.; Dunlop, J.; Dalziel, J. E. Biosensors and Bioelectronics 2007,22, 1006 ? 1012.[16] Sakai, N.; Sord?, N.; Matile, S. Journal of the American Chemical Society2003, 125, 7776?7777.[17] Cornell, B. A.; Braach-Maksvytis, V. L. B.; King, L. G.; Osman, P. D. J.;Raguse, B.; Wieczorek, L.; Pace, R. J. Nature 1997, 387, 580?583.[18] Oh, S.; Cornell, B.; Smith, D.; Higgins, G.; Burrell, C.; Kok, T. Biosensorsand Bioelectronics 2008, 23, 1161 ? 1165.[19] Naumann, R.; Walz, D.; Schiller, S.; Knoll, W. Journal of ElectroanalyticalChemistry 2003, 550-551, 241 ? 252.[20] Steinem, C.; Galla, H.-J.; Janshoff, A. Phys. Chem. Chem. Phys. 2000, 2,4580?4585.[21] Steinem, C.; Janshoff, A.; Galla, H.-J.; Sieber, M. Bioelectrochemistry andBioenergetics 1997, 42, 213 ? 220.[22] Gritsch, S.; Nollert, P.; J?hnig, F.; Sackmann, E. Langmuir 1998, 14, 3118?3125.[23] Knapp, O.; Benz, R.; Gibert, M.; Marvaud, J. C.; Popoff, M. R. Journal ofBiological Chemistry 2002, 277, 6143?6152.[24] Hirano, A.; Wakabayashi, M.; Matsuno, Y.; Sugawara, M. Biosensors andBioelectronics 2003, 18, 973 ? 983.[25] Glazier, S. A.; Vanderah, D. J.; Plant, A. L.; Bayley, H.; Valincius, G.;Kasianowicz, J. J. Langmuir 2000, 16, 10428?10435.[26] Grieshaber, D.; MacKenzie, R.; V?r?s, J.; Reimhult, E. Sensors 2008, 8,1400?1458.177References[27] Jesorka, A.; Orwar, O. Annu. Rev. Anal. Chem. 2008, 1, 801?832.[28] Wilkinson, D.; Nagle, J. In Liposomes: From Physical Structure to Ther-apeutic Applications; Knight, C., Ed.; Elsevier, 1981; Vol. 7; Chapter 9 -Thermodynamics of Lipid Bilayers, pp 273?294.[29] Sackmann, E. In Handbook of Biological Physics; Lipowsky, R., Sack-mann, E., Eds.; Elsevier, 1995; Vol. 1; Chapter Physical Basis of Self-Organization and Function of Membranes: Physics of Vesicles, pp 213?303.[30] Pockels, A. Nature 1891, 43, 437?439.[31] Blodgett, K. Journal of the American Chemical Society 1934, 56, 495.[32] Langmuir, I. Journal of the American Chemical Society 1917, 39, 1848?1906.[33] Gaines, G. L. In Insoluble Monolayers at Gas-Liquid Interfaces; Pri-gogine, I., Ed.; Interscience Monographs on Chemistry; John Wiley & SonsLtd., 1966; p 386.[34] Harkins, W. D.; Copeland, L. E. The Journal of Chemical Physics 1942, 10,272?286.[35] Lawrie, G.; Barnes, G. Journal of Colloid and Interface Science 1994, 162,36 ? 44.[36] Bibo, A. M.; Knobler, C. M.; Peterson, I. R. The Journal of Physical Chem-istry 1991, 95, 5591?5599.[37] Brezesinski, G.; Kaganer, V. M.; Mohwald, H.; Howes, P. B. The Journal ofChemical Physics 1998, 109, 2006?2010.[38] Steitz, R.; Peng, J. B.; Peterson, I. R.; Gentle, I. R.; Kenn, R. M.; Gold-mann, M.; Barnes, G. T. Langmuir 1998, 14, 7245?7249.[39] Slevin, C. J.; Ryley, S.; Walton, D. J.; Unwin, P. R. Langmuir 1998, 14,5331?5334.[40] Kwok, D. Y.; Tadros, B.; Deol, H.; Vollhardt, D.; Miller, R.; Cabrerizo-V?-lchez, M. A.; Neumann, A. W. Langmuir 1996, 12, 1851?1859.178References[41] Petty, M. C. Langmuir-Blodgett films: an introduction; Cambridge Univer-sity Press, 1996; Online resource.[42] Zdravkova, A.; van der Eerden, J.; Snel, M. Journal of Crystal Growth 2005,275, e1029 ? e1033.[43] Kuramori, M.; Ishikawa, T.; Narita, T.; Oishi, Y. Progress in Colloid andPolymer Science 2011, 138, 103?107.[44] Winterhalter, M. Current Opinion in Colloid & Interface Science 2000, 5,250 ? 255.[45] Atkins, P. W. Physical Chemistry; Oxford University press, 1978; ISBN 0-7167-1381-0.[46] Zawisza, I.; Lipkowski, J. Langmuir 2004, 20, 4579?4589.[47] Bilewicz, R.; Majda, M. Langmuir 1991, 7, 2794?2802.[48] Sagiv, J. Journal of the American Chemical Society 1980, 102, 92?98.[49] Chen, D.; Li, J. Surface Science Reports 2006, 61, 445 ? 463.[50] Ajami, E.; Aguey-Zinsou, K.-F. Journal of Colloid and Interface Science2012, 385, 258 ? 267.[51] Allongue, P.; de Villeneuve, C. H.; Cherouvrier, G.; Cort?s, R.; Bernard, M.-C. Journal of Electroanalytical Chemistry 2003, 550-551, 161?174.[52] Mandler, D.; Kraus-Ophir, S. Journal of Solid State Electrochemistry 2011,15, 1535?1558.[53] Weisshaar, D. E.; Lamp, B. D.; Porter, M. D. Journal of the American Chem-ical Society 1992, 114, 5860?5862.[54] Uvdal, K.; Vikinge, T. P. Langmuir 2001, 17, 2008?2012.[55] Chen, S.; Li, L.; Boozer, C. L.; Jiang, S. Langmuir 2000, 16, 9287?9293.[56] Ottova, A.; Tvarozek, V.; Tien, H. In Planar Lipid Bilayers (BLMs) and theirApplications; Tien, H., Ottova-Leitmannova, A., Eds.; Elsevier Science B.V.,2003; Chapter Supported Planar Lipid Bilayers, pp 917?961.179References[57] Brian, A. A.; McConnell, H. M. Proceedings of the National Academy ofSciences of the United States of America 1984, 81, pp. 6159?6163.[58] Hamai, C.; Yang, T.; Kataoka, S.; Cremer, P. S.; Musser, S. M. BiophysicalJournal 2006, 90, 1241 ? 1248.[59] Johnson, J. M.; Ha, T.; Chu, S.; Boxer, S. G. Biophysical Journal 2002, 83,3371 ? 3379.[60] Keller, C.; Kasemo, B. Biophys. J . 1998, 75, 1397?1402.[61] Reimhult, E.; Hook, F.; Kasemo, B. Langmuir 2003, 19, 1681?1691.[62] Wu, H.-L.; Chen, P.-Y.; Chi, C.-L.; Tsao, H.-K.; Sheng, Y.-J. Soft Matter2013, 9, 1908?1919.[63] Mui, B.; Cullis, P.; Evans, E.; Madden, T. Biophysical Journal 1993, 64, 443? 453.[64] Sugihara, K.; Jang, B.; Schneider, M.; Voros, J.; Zambelli, T. Soft Matter2012, 8, 5525?5531.[65] Xu, S.; Szymanski, G.; Lipkowski, J. JACS 2004, 126, 12276?12277.[66] Sek, S.; Xu, S.; Chen, M.; Szymanski, G.; Lipkowski, J. JACS 2008, 130,5736?5743.[67] Agak, J. O.; Stoodley, R.; Retter, U.; Bizzotto, D. J. Electroanal. Chem.2004, 562, 135 ? 144.[68] Agmo Hern?ndez, V.; Scholz, F. Langmuir 2006, 22, 10723?10731.[69] Stoodley, R.; Bizzotto, D. Analyst 2003, 128, 552?561.[70] Zan, G. H.; Tan, C.; Deserno, M.; Lanni, F.; Losche, M. Soft Matter 2012,8, 10877?10886.[71] Agmo Hern?ndez, V. Journal of Solid State Electrochemistry 2013, 17, 299?305.[72] Agmo Hern?ndez, V.; Milchev, A.; Scholz, F. Journal of Solid State Electro-chemistry 2009, 13, 1111?1114.180References[73] McConnell, H.; Watts, T.; Weis, R.; Brian, A. Biochimica et Biophysica Acta(BBA) - Reviews on Biomembranes 1986, 864, 95 ? 106.[74] Albertorio, F.; Diaz, A. J.; Yang, T.; Chapa, V. A.; Kataoka, S.; Castel-lana, E. T.; Cremer, P. S. Langmuir 2005, 21, 7476?7482, PMID: 16042482.[75] Castellana, E. T.; Cremer, P. S. Surf. Sci. Rep. 2006, 61, 429 ? 444.[76] Williams, L. M.; Evans, S. D.; Flynn, T. M.; Marsh, A.; Knowles, P. F.;Bushby, R. J.; Boden, N. Langmuir 1997, 13, 751?757.[77] Giess, F.; Friedrich, M. G.; Heberle, J.; Naumann, R. L.; Knoll, W. Biophys-ical Journal 2004, 87, 3213 ? 3220.[78] Knoll, W.; Frank, C.; Heibel, C.; Naumann, R.; Offenh?usser, A.; R?he, J.;Schmidt, E.; Shen, W.; Sinner, A. Reviews in Molecular Biotechnology 2000,74, 137 ? 158.[79] Knoll, W.; K?per, I.; Naumann, R.; Sinner, E.-K. Electrochimica Acta 2008,53, 6680 ? 6689.[80] Inaba, T.; Tatsu, Y.; Morigaki, K. Langmuir 2011, 27, 12515?12520.[81] Groves, J. T.; Ulman, N.; Boxer, S. G. Science 1997, 275, 651?653.[82] Groves, J. T.; Ulman, N.; Cremer, P. S.; Boxer, S. G. Langmuir 1998, 14,3347?3350.[83] Cremer, P. S.; Yang, T. JACS 1999, 121, 8130?8131.[84] Groves, J. T.; Mahal, L. K.; Bertozzi, C. R. Langmuir 2001, 17, 5129?5133.[85] Hovis, J. S.; Boxer, S. G. Langmuir 2000, 16, 894?897.[86] Sekula, S.; Fuchs, J.; Weg-Remers, S.; Nagel, P.; Schuppler, S.; Fragala, J.;Theilacker, N.; Franzreb, M.; Wingren, C.; Ellmark, P.; Borrebaeck, C.A. K.; Mirkin, C. A.; Fuchs, H.; Lenhert, S. Small 2008, 4, 1785?1793.[87] Yee, C.; Amweg, M.; Parikh, A. Advanced Materials 2004, 16, 1184?1189.[88] Orth, R.; Kameoka, J.; Zipfel, W.; Ilic, B.; Webb, W.; Clark, T.; Craig-head, H. Biophysical Journal 2003, 85, 3066 ? 3073.181References[89] Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals andApplications, 2nd ed.; John Wiley & Sons Ltd., 2001.[90] Ross, P. N.; D?Agostino, A. T. Electrochimica Acta 1992, 37, 615 ? 623.[91] Kolb, D.; Lehmpfuhl, G.; Zei, M. Journal of Electroanalytical Chemistryand Interfacial Electrochemistry 1984, 179, 289 ? 295.[92] Hamelin, A. Journal of Electroanalytical Chemistry and Interfacial Electro-chemistry 1982, 142, 299 ? 316.[93] Macdonald, J. R.; Carl A. Barlow, J. The Journal of Chemical Physics 1962,36, 3062?3080.[94] Damaskin, B.; Petrii, O. Journal of Solid State Electrochemistry 2011, 15,1317?1334.[95] Buess-Herman, C. Journal of Electroanalytical Chemistry and InterfacialElectrochemistry 1985, 186, 27 ? 39.[96] Baikerikar, K.; Sathyanarayana, S. Journal of Electroanalytical Chemistryand Interfacial Electrochemistry 1970, 24, 333 ? 344.[97] Buess-Herman, C.; Franck, C.; Gierst, L. Journal of ElectroanalyticalChemistry 1992, 329, 91 ? 102.[98] Lipkowski, J.; Stolberg, L. In Adsorption of Molecules at Metal Electrodes;Lipkowski, J., Ross, P. N., Eds.; Frontiers of Electrochemistry; VCH Pub-lishers Inc., 1992; Chapter Molecular Adsorption at Gold and Silver Elec-trodes, pp 171?238.[99] Damaskin, B. Russian Journal of Electrochemistry 2007, 43, 9?16.[100] Grahame, D. C. Chemical Reviews 1947, 41, 441?501.[101] Buess-Herman, C. In Adsorption of Molecules at Metal Electrodes; Lip-kowski, J., Ross, P. N., Eds.; Frontiers of Electrochemistry; VCH PublishersInc., 1992; Chapter Dynamics of Adsorption and Two-Dimensional PhaseTransitions at Electrode Surfaces, pp 77 ? 118.182References[102] Parsons, R. In Advances in Electrochemistry and Electrochemical Engineer-ing; Delahay, P., Ed.; Interscience Publishers, Inc., 1961; Vol. 1; Chapter 1,pp 2?61.[103] Jovic?, V.; Jovic?, B. Journal of Electroanalytical Chemistry 2003, 541, 1 ?11.[104] Ralstrick, I. D.; Macdonald, J. R.; Franceschetti, D. R. In Impedance Spec-troscopy, Emphasizing Solid Materials and Systems; Macdonald, J. R., Ed.;John Wiley & Sons Ltd., 1987; Chapter Theory, pp 27?132.[105] Pajkossy, T. Solid State Ionics 1997, 94, 123 ? 129.[106] Bizzotto, D. Characterization of the adsorption of insoluble surfactants ontoan electrified interface. Ph.D. thesis, The University of Guelph, 1996.[107] Bizzotto, D.; Lipkowski, J. In Amphiphiles at Interfaces; Texter, J., Ed.;Progress in Colloid & Polymer Science; Steinkopff, 1997; Vol. 103; pp 201?215.[108] Dickinson, K. M.; Hanson, K. E.; Fredlein, R. A. Electrochimica Acta 1992,37, 139 ? 141.[109] Yang, Y.; Bizzotto, D. J. Electroanal. Chem. 2001, 500, 408 ? 417.[110] Bizzotto, D.; No?l, J. J.; Lipkowski, J. Thin Solid Films 1994, 248, 69?77.[111] Bizzotto, D.; Wong, E.; Yang, Y. Journal of Electroanalytical Chemistry2000, 480, 233 ? 240.[112] Zawisza, I.; Burgess, I.; Szymanski, G.; Lipkowski, J.; Majewski, J.;Satija, S. Electrochim. Acta 2004, 49, 3651 ? 3664, The role of electro-chemistry in the sustained development of modern societies.[113] Bizzotto, D.; Yang, Y.; Shepherd, J. L.; Stoodley, R.; Agak, J.; Stauffer, V.;Lathuilliere, M.; Akhtar, A. S.; Chung, E. J. Electroanal. Chem. 2004, 574,167 ? 184.[114] Shepherd, J. L. Electro-fluorescence Characterization of Insoluble Surfac-tants Adsorbed on Solid Electrodes. Ph.D. thesis, University of BritishColumbia, 2005.183References[115] Bizzotto, D.; Pettinger, B. Langmuir 1999, 15, 8309?8314.[116] Miller, I. R.; Bach, D. Journal of Colloid and Interface Science 1969, 29,250 ? 260.[117] Bach, D.; Miller, I.; Robert, F. Chemistry and Physics of Lipids 1970, 4, 269? 288.[118] Bach, D.; Miller, I. Electrochimica Acta 1970, 15, 533 ? 543.[119] Miller, I.; Rishpon, J.; Tenenbaum, A. Bioelectrochemistry and Bioenerget-ics 1976, 3, 528 ? 542.[120] Stoodley, R.; Shepherd, J.; Wasan, K.; Bizzotto, D. Biochimica et BiophysicaActa (BBA) - Biomembranes 2002, 1564, 289 ? 297.[121] Nelson, A.; Auffret, N.; Borlakoglu, J. Biochimica et Biophysica Acta (BBA)- Biomembranes 1990, 1021, 205 ? 216.[122] Nelson, A.; Auffret, N.; Readman, J. Analytica Chimica Acta 1988, 207, 47? 57.[123] Bizzotto, D.; Nelson, A. Langmuir 1998, 14, 6269?6273.[124] Stauffer, V.; Stoodley, R.; Agak, J. O.; Bizzotto, D. J. Electroanal. Chem.2001, 516, 73 ? 82.[125] Agak, J. O. The Influence of Cations on DOPC Adsorbed ontoHg/Electrolyte Interface Electrochemical and Impedance SpectroscopyStudy. M.Sc. thesis, The University of British Columbia, 2002.[126] Zawisza, I.; Bin, X.; Lipkowski, J. Bioelectrochemistry 2004, 63, 137 ? 147.[127] Hoffmannov?, H.; Hof, M.; Krtil, P. J. Electroanal. Chem. 2006, 588, 296 ?302.[128] Lindholm-Sethson, B.; Geladi, P.; Nelson, A. Analytica Chimica Acta 2001,446, 121 ? 130.[129] Burgess, I.; Li, M.; Horswell, S.; Szymanski, G.; Lipkowski, J.; Majew-ski, J.; Satija, S. Biophys. J . 2004, 86, 1763?1776.184References[130] Burgess, I.; Li, M.; Horswell, S.; Szymanski, G.; Lipkowski, J.; Satija, S.;Majewski, J. Colloids Surf., B 2005, 40, 117 ? 122.[131] Horswell, S. L.; Zamlynny, V.; Li, H.-Q.; Merrill, A. R.; Lipkowski, J. Fara-day Discuss. 2002, 121, 405 ? 422.[132] Zawisza, I.; Lachenwitzer, A.; Zamlynny, V.; Horswell, S.; Goddard, J.; Lip-kowski, J. Biophys. J . 2003, 85, 4055?4075.[133] Bin, X.; Zawisza, I.; Goddard, J. D.; Lipkowski, J. Langmuir 2005, 21, 330?347.[134] Sinner, A.; Offenhausser, A. Thin Solid Films 1998, 327-329, 758 ? 761.[135] Ali, M. H.; Kirby, D. J.; Mohammed, A. R.; Perrie, Y. Journal of Pharmacyand Pharmacology 2010, 62, 1646?1655.[136] DeNardis, N. I.; ?utic?, V.; Svetlic?ic?, V.; Frkanec, R. The Journal of Mem-brane Biology 2012, 245, 573?582.[137] Tsekov, R.; Kovacic?, S.; ?utic?, V. Langmuir 1999, 15, 5649?5653.[138] Tomaic?, J.; Legovic?, T.; ?utic?, V. Journal of Electroanalytical Chemistry1992, 322, 79 ? 92.[139] Kovac?, S.; Svetlic?ic?, V.; ?utic?, V. Colloids and Surfaces A: Physicochemicaland Engineering Aspects 1999, 149, 481 ? 489.[140] Svetlic?ic?, V.; Ivo?evic?, N.; Kovac?, S.; ?utic?, V. Langmuir 2000, 16, 8217?8220.[141] ?utic?, V.; Kovac?, S.; Tomaic?, J.; Svetlic?ic?, V. Journal of ElectroanalyticalChemistry 1993, 349, 173 ? 186.[142] Dimova, R.; Aranda, S.; Bezlyepkina, N.; Nikolov, V.; Riske, K. A.;Lipowsky, R. J. Phys. Condens. Matter 2006, 18, S1151?S1176.[143] Mitov, M. D.; M?l?ard, P.; Winterhalter, M.; Angelova, M. I.; Bothorel, P.Phys. Rev. E 1993, 48, 628?631.[144] Correa, N. M.; Schelly, Z. A. The Journal of Physical Chemistry B 1998,102, 9319?9322.185References[145] Lim, J. K.; Zhou, H.; Tilton, R. D. J. Colloid Interface Sci. 2009, 332, 113 ?121.[146] Lakowicz, J. R. In Principles of Fluorescence Spectroscopy; Lakowicz, J. R.,Ed.; Springer US, 2006; Chapter 13: Energy Transfer, pp 443?475.[147] Deniz, A. A.; Laurence, T. A.; Beligere, G. S.; Dahan, M.; Martin, A. B.;Chemla, D. S.; Dawson, P. E.; Schultz, P. G.; Weiss, S. Proceedings of theNational Academy of Sciences 2000, 97, 5179?5184.[148] Jia, Y.; Talaga, D. S.; Lau, W. L.; Lu, H. S.; DeGrado, W. F.;Hochstrasser, R. M. Chemical Physics 1999, 247, 69 ? 83.[149] Schuler, B.; Eaton, W. A. Current Opinion in Structural Biology 2008, 18,16 ? 26.[150] Tsuji, A.; Koshimoto, H.; Sato, Y.; Hirano, M.; Sei-Iida, Y.; Kondo, S.;Ishibashi, K. Biophysical Journal 2000, 78, 3260 ? 3274.[151] Juskowiak, B. Analytica Chimica Acta 2006, 568, 171 ? 180.[152] Wang, X.; Lou, X.; Wang, Y.; Guo, Q.; Fang, Z.; Zhong, X.; Mao, H.; Jin, Q.;Wu, L.; Zhao, H.; Zhao, J. Biosensors and Bioelectronics 2010, 25, 1934 ?1940.[153] Hsieh, A.-H.; Pan, P.-H.; Lee, A. Microfluidics and Nanofluidics 2009, 6,391?401.[154] Zacharias, D. A.; Violin, J. D.; Newton, A. C.; Tsien, R. Y. Science 2002,296, 913?916.[155] Glebov, O. O.; Nichols, B. J. Nat Cell Biol 2004, 6, 238?243.[156] Loura, L. M.; de Almeida, R. F.; Silva, L. C.; Prieto, M. Biochimica et Bio-physica Acta (BBA) - Biomembranes 2009, 1788, 209 ? 224.[157] Sengupta, P.; Holowka, D.; Baird, B. Biophysical Journal 2007, 92, 3564 ?3574.[158] Mikhalyov, I.; Gretskaya, N.; Bergstrom, F.; Johansson, L. B.-A. Phys.Chem. Chem. Phys. 2002, 4, 5663?5670.186References[159] Tleugabulova, D.; Zhang, Z.; Brennan, J. D. The Journal of Physical Chem-istry B 2002, 106, 13133?13138.[160] Bergstr?m, F.; Mikhalyov, I.; H?ggl?f, P.; Wortmann, R.; Ny, T.; Johans-son, L. B.-o. Journal of the American Chemical Society 2002, 124, 196?204,PMID: 11782171.[161] Casanova-Moreno, J. R.; Bizzotto, D. Langmuir 2013, 29, 2065?2074.[162] Casanova-Moreno, J.; Bizzotto, D. J. Electroanal. Chem. 2010, 649, 126?135.[163] Musgrove, A.; Kell, A.; Bizzotto, D. Langmuir 2008, 24, 7881?7888.[164] Shepherd, J. L.; Kell, A.; Chung, E.; Sinclar, C. W.; Workentin, M. S.; Biz-zotto, D. JACS 2004, 126, 8329?8335.[165] Janosch, S.; Nicolini, C.; Ludolph, B.; Peters, C.; V?lkert, M.; Hazlet, T. L.;Gratton, E.; Waldmann, H.; Winter, R. Journal of the American ChemicalSociety 2004, 126, 7496?7503, PMID: 15198596.[166] Ziessel, R.; Harriman, A. Chem. Commun. 2011, 47, 611?631.[167] Drexhage, K. Journal of Luminescence 1970, 1-2, 693 ? 701.[168] Chance, R. R.; Miller, A. H.; Prock, A.; Silbey, R. The Journal of ChemicalPhysics 1975, 63, 1589?1595.[169] Das, P.; Metiu, H. The Journal of Physical Chemistry 1985, 89, 4680?4687.[170] Weitz, D. A.; Garoff, S.; Hanson, C. D.; Gramila, T. J.; Gersten, J. I. Opt.Lett. 1982, 7, 89?91.[171] Campion, A.; Gallo, A.; Harris, C.; Robota, H.; Whitmore, P. ChemicalPhysics Letters 1980, 73, 447 ? 450.[172] Lakowicz, J. R. Analytical Biochemistry 2005, 337, 171 ? 194.[173] Lakowicz, J. R.; Shen, Y.; D?Auria, S.; Malicka, J.; Fang, J.; Gryczynski, Z.;Gryczynski, I. Analytical Biochemistry 2002, 301, 261 ? 277.[174] Amos, R. M.; Barnes, W. L. Phys. Rev. B 1999, 59, 7708?7714.187References[175] Axelrod, D.; Hellen, E. H.; Fulbright, R. M. In Topics in Fluorescence Spec-troscopy, Volume 3: Biochemical Applications; Lakowicz, J. R., Ed.; KluwerAcademic Publishers, 1992; Chapter 7: Total Internal Reflection Fluores-cence, pp 289?343.[176] Rant, U.; Arinaga, K.; Fujita, S.; Yokoyama, N.; Abstreiter, G.; Tornow, M.Nano Letters 2004, 4, 2441?2445.[177] Rant, U.; Arinaga, K.; Scherer, S.; Pringsheim, E.; Fujita, S.; Yokoyama, N.;Tornow, M.; Abstreiter, G. Proceedings of the National Academy of Sciences2007, 104, 17364?17369.[178] Murphy, D. B.; Davidson, M. W. Fundamentals of Light Microscopy andElectronic Imaging; Wiley-Blackwell, 2012.[179] Quercioli, F. In Optical Fluorescence Microscopy; Diaspro, A., Ed.;Springer-Verlag, 2011; Chapter 1: Fundamentals of Optical Microscopy, pp1?36.[180] Bizzotto, D.; Lipkowski, J. Journal of Electroanalytical Chemistry 1996,409, 33 ? 43.[181] Li,; Ruzgas, T.; Gaigalas, A. K. Langmuir 1999, 15, 6358?6363.[182] Li, L.; Meuse, C.; Silin, V.; Gaigalas, A. K.; Zhang, Y.-z. Langmuir 2000,16, 4672?4677.[183] Shepherd, J. L.; Bizzotto, D. Langmuir 2006, 22, 4869?4876.[184] Shepherd, J. L.; Bizzotto, D. The Journal of Physical Chemistry B 2003,107, 8524?8531.[185] Shepherd, J.; Yang, Y.; Bizzotto, D. J. Electroanal. Chem. 2002, 524-525,54 ? 61.[186] Haugstad, G. Atomic Force Microscopy: Understanding Basic Modes andAdvanced Applications; John Wiley & Sons Ltd., 2012.[187] K?nig, T.; Papke, T.; Kopyshev, A.; Santer, S. Journal of Materials Science2013, 48, 3863?3869.188References[188] Kim, Y.; Lieber, C. M. Science 1992, 257, 375?377.[189] Paolicelli, G.; Mougin, K.; Vanossi, A.; Valeri, S. Journal of Physics: Con-densed Matter 2008, 20, 354011.[190] Han, W. Ultramicroscopy 2008, 108, 1009 ? 1012, Proceedings of the NinthInternational Conference on Scanning Probe Micrososcopy, Sensors andNanostructures.[191] Han, W.; Lindsay, S. M.; Jing, T. Appl. Phys. Lett. 1996, 69, 4111?4113.[192] Han, X.; Pradeep, S.; Critchley, K.; Sheikh, K.; Bushby, R.; Evans, S.Chemistry-A European Journal 2007, 13, 7957?7964.[193] Kienberger, F.; Stroh, C.; Kada, G.; Moser, R.; Baumgartner, W.; Pas-tushenko, V.; Rankl, C.; Schmidt, U.; M?ller, H.; Orlova, E.; LeGrimel-lec, C.; Drenckhahn, D.; Blaas, D.; Hinterdorfer, P. Ultramicroscopy 2003,97, 229 ? 237.[194] Vermette, P. In Methods in Enzymology; Duzgunes, N., Ed.; Methods inEnzymology; Academic Press, 2009; Vol. 465; pp 43 ? 73.[195] Li, X.-F.; Zhang, G.-Y.; Dong, J.-F.; Zhou, X.-H.; Hong, X.-L. Chinese Jour-nal of Chemistry 2006, 24, 311?315.[196] Sch?nherr, H.; Johnson, J. M.; Lenz, P.; Frank, C. W.; Boxer, S. G. Langmuir2004, 20, 11600?11606, PMID: 15595789.[197] Egawa, H.; Furusawa, K. Langmuir 1999, 15, 1660?1666.[198] Nakano, K.; Tozuka, Y.; Yamamoto, H.; Kawashima, Y.; Takeuchi, H. Int. J.Pharm. 2008, 355, 203 ? 209.[199] Liang, X.; Mao, G.; Ng, K. S. Colloids and Surfaces B: Biointerfaces 2004,34, 41 ? 51.[200] Liang, X.; Mao, G.; Ng, K. Journal of Colloid and Interface Science 2004,278, 53 ? 62.[201] Lawrence, J. C.; Saslowsky, D. E.; Edwardson, J. M.; Henderson, R. M.Biophysical Journal 2003, 84, 1827?1832.189References[202] Johnston, L. J. Langmuir 2007, 23, 5886?5895.[203] Orsini, F.; Cremona, A.; Arosio, P.; Corsetto, P.; Montorfano, G.; Lascial-fari, A.; Rizzo, A. Biochimica et Biophysica Acta (BBA) - Biomembranes2012, 1818, 2943 ? 2949.[204] Yuan, C.; Furlong, J.; Burgos, P.; ; Johnston, L. J. Biophysical Journal 2002,82, 2526.[205] Tokumasu, F.; Jin, A. J.; Feigenson, G. W.; Dvorak, J. A. Ultramicroscopy2003, 97, 217?27.[206] Shaw, J. E.; Epand, R. F.; Epand, R. M.; Li, Z.; Bittman, R.; Yip, C. M.Biophys. J. 2006, 90, 2170?2178.[207] Xu, S.; Chen, M.; Cholewa, E.; Szymanski, G.; Lipkowski, J. Langmuir2007, 23, 6937?6946.[208] Cunha, F.; Tao, N. J.; Wang, X. W.; Jin, Q.; Duong, B.; D?Agnese, J. Lang-muir 1996, 12, 6410?6418.[209] Kelley, S. O.; Barton, J. K.; Jackson, N. M.; McPherson, L. D.; Potter, A. B.;Spain, E. M.; Allen, M. J.; Hill, M. G. Langmuir 1998, 14, 6781?6784.[210] Li, M.; Chen, M.; Sheepwash, E.; Brosseau, C. L.; Li, H.; Pettinger, B.;Gruler, H.; Lipkowski, J. Langmuir 2008, 24, 10313?10323, PMID:18712889.[211] Mayer, L.; Hope, M.; Cullis, P. Biochim. Biophys. Acta 1986, 858, 161?168.[212] National Institutions of Health, ImageJ 1.47d.[213] Russ, J. C. The Image Processing Handbook, 2nd ed.; CRC Press, Inc: BocaRaton, Florida, 1994; pp 253?254.[214] Czech Metrology Institute, Gwyddion 2.31.[215] Carter Ramirez, D. M.; Pitre, S. P.; Kim, Y. A.; Bittman, R.; Johnston, L. J.Langmuir 2013, 29, 3380?3387.190[216] Ira,; Zou, S.; Ramirez, D. M. C.; Vanderlip, S.; Ogilvie, W.; Jakubek, Z. J.;Johnston, L. J. Journal of Structural Biology 2009, 168, 78 ? 89.[217] Spencer Kimball, Peter Mattis and the GIMP Development Team, GNU Im-age Manipulation Program (GIMP) 2.8.[218] Phung, T.; Zhang, Y.; Dunlop, J.; Dalziel, J. E. International Journal ofElectrochemistry 2011, 2011, 213107.[219] Braha, O.; Walker, B.; Cheley, S.; Kasianowicz, J. J.; Song, L.; Gouaux, J.;Bayley, H. Chemistry & Biology 1997, 4, 497 ? 505.[220] Grit, M.; Crommelin, D. J. Chemistry and Physics of Lipids 1993, 64, 3 ?18.[221] Marathe, D.; Mishra, K. P. Radiat. Res. 2002, 157, pp. 685?692.[222] Panasenko, O. M.; Arnhold, J.; Sergienko, V. I. Membrane and Cell Biology1997, 11, 253?8.[223] Sternberg, S. R. Computer 1983, 16, 22?34.191Appendix ALiposome Characterization andStabilityA.1 Characterization by Thin Layer ChromatographyPhospholipids, such as the DOPC used in these studies, are subject to degradationover time. In aqueous solution, the carboxy esters linking the fatty acid residuesto the glycerol backbone may be hydrolyzed to release the fatty acids, eventuallyproducing glycerophosphoric acid if both fatty acid chains are hydrolyzed. [220]Thin layer chromatography (TLC) can be used to identify liposomal suspensionsin which the phospholipid has degraded and used as a screening method to ensurethat solutions used are not decomposed. [221]A.1.1 Experimental MethodsLiposome solutions were spotted onto TLC plates (1 ?L) onto an aluminum-backedsilica gel TLC plate (EMD) along with a sample of stock DOPC solution in chlo-roform (0.5 ?L) and a known degraded liposome solution (1 ?L) in adjacent lanes.The degraded liposome solution was prepared from liposomes extruded in the per-chlorate electrolyte, saturated with oxygen after preparation, and aged for severalweeks. After applying the solution to the TLC plate, the spots were allowed to drythoroughly and then eluted using a 2:1 chloroform:methanol mixture (both: Fisher,HPLC grade). After elution, the plate was dried and visualized by rinsing withdilute sulphuric acid (Fisher ACS, approximately 1 M) and charring in an oven at190?C for 5-10 minutes until spots appeared. Solutions were monitored in this man-ner from formation until the end of use of the solution. During the characterizationphase, liposomes were initially measured every few days. After it was establishedthat the liposomes in the NaF electrolyte were more stable, confirmation of purityby TLC was performed every few weeks.192A.2. Characterization by Elastic Light ScatteringA.1.2 ResultsSelected TLC plates from the monitoring of Solution 1 of liposomes are shown inFigure A.1. For clarity in viewing, the plates have been electronically scanned, thecolor desaturated, and the brightness-contrast adjusted to enhance visibility. Spotscorresponding to the fatty acid degradation product are marked with a black circle.Two solutions of liposomes prepared on the same day, one in 0.1 M NaF and onein 0.05 M KClO4, are compared with a known degraded liposome solution and astock solution of DOPC in chloroform. It is easily seen that in just a few weeks,the liposomes in the KClO4 electrolyte have begun to degrade, producing a secondspot on the TLC plate. After nearly 19 months of storage and use, the final TLCtaken of the solution before it was used up shows no evidence of degradation in theNaF electrolyte, while the second spot in the KClO4 electrolyte solution has growndarker.The instability of liposomes in the 0.05 M KClO4 electrolyte is not unexpected,as perchlorate is a known oxidizer and would be expected to enhance the hydroly-sis of the ester bonds. A previous study [222] suggested that perchlorate does notcontribute to peroxidization of the phospholipids, however other mechanisms ofdegradation are possible and are clearly occurring in storage. For this reason, stud-ies done with liposomes prepared in perchlorate electrolyte (for example [127])should be considered carefully and caution taken if work is done on liposome sus-pensions that were not freshly prepared.A.2 Characterization by Elastic Light ScatteringIn order to confirm that the liposomes produced were of uniform and correct size,an initial characterization of the liposome solution was performed by dynamic lightscattering. A Coulter N4+ particle size analyzer was used, and sizes taken as an av-erage of 3 measurements each integrated over 120 s. The sample was 5 ?L ofthe liposome solution diluted to approximately 4 mL using the same 0.1 M NaFelectrolyte in which the liposomes were prepared. Temperature was held at 20?C,equilibrated for 5 min, and scattering measured at 90?. For each of the three mea-surements, a particle size histogram from 1 - 1000 nm was recorded (31 bins) aswell as a size histogram from 50 - 250 nm (31 bins). The larger range ensuresthat there are no particles measured well outside the expected range, while the sec-193A.2. Characterization by Elastic Light ScatteringFigure A.1: TLC plates taken at various points during the lifetime of a liposomesolution. From left to right, the plates were measured on: Oct 25 2010 (solution age0 days), Nov 2 2010 (solution age 8 days), Dec 20 2010 (solution age 8 weeks), andMay 25 2011 (solution age 7 months - final use). All plates are spotted with Lane 1- Liposome solution in NaF, Lane 2 - Liposome solution in KClO4, Lane 3 - Knowndegraded liposome solution, Lane 4 - DOPC stock solution in chloroform. Lanes1-3 have a 1 ?L spot, Lane 4 is 0.5 ?L. Spots with a lower retention factor (closerto the solvent front) are the fatty acid degradation product and are highlighted withblack circles. In Lane 3 and the final measurement of Lane 2, the higher retentionfactor spot from the lysophospholipid degradation product is also visible. Identityof the markings along the solvent front is unknown but is present in blank runs.ond histogram provides greater detail on the size distribution within the maximumnumber of bins available. The data from the histogram was fit to a log-normal dis-tribution using QTIplot (Ion Vasilief, the mean particle size calculated and averaged for the three measurements. Thedata for the two liposome preparations used experimentally is shown in Table A.1.As the nominal particle size was 100 nm based on the pore size of the filter used,the size was deemed acceptable if it was roughly similar. Both preparations wereconsidered acceptably close to 100 nm.194A.2. Characterization by Elastic Light ScatteringPreparation MeanSize(nm)Standard Deviation(nm)Solution 1Feb. 2011 133 15Solution 2Aug 2012 122 20Table A.1: Mean particle sizes for liposome solutions as measured using dynamiclight scattering.195Appendix BFluorescence Calibration FactorsCorrection factors for comparing the two exposure time / gain settings used wasdone empirically. Based on measurements of a calibration sample (A3 on Fo-calCheck Test Slide #1, Molecular Probes), the average fluorescence intensity wasfound to be directly proportional to the exposure time with the same electron-multiplier gain settings, and revealed a dark signal of 500. The conversion factorto kcts/sec for matching the images acquired at 2.5 sec and 200 gain to the 5 sec,400 gain images was determined by dividing the average grayscale values of sev-eral images taken at these settings, and testing the conversion factor against furtherimages that were not a part of the analysis. The conversion factor was found to berelated to the exposure time and gain ratio, but not perfectly. The values used in theanalyses presented in this work are summarized in Table B.1.Image Settings Conversion Factorto kcts/secExposure: 2.5 secE.M. gain: 200 2500Exposure: 5 secE.M. gain: 400 7500Table B.1: Conversion factors to kcts/sec used for fluorescence images.196Appendix CRolling-Ball Image ProcessingIn Chapters 5 and 6, a rolling-ball filter is used to remove background intensitygradients from fluorescence images while maintaining local changes in image in-tensity. The rolling-ball filter, also called a ?top-hat? filter, can be imagined asrolling a ball of diameter X over the image, where the intensity value is mapped asheight. [223] An average intensity value is taken over the radius of the ball, and thisbackground value is subtracted from that point. If the radius is sufficiently large,it will remove low-frequency background fluctuations as well as the average back-ground intensity while leaving unaltered features with a smaller radius. Choosingan appropriate radius is important as a too small radius will result in loss of someimage features. For comparison, in Figure C.1 a fluorescence image from Chapter5 is processed using three different ball sizes.197Appendix C. Rolling-Ball Image Processing(a) 5-px ball (b) 50-px ball(c) 250-px ball (d) Raw (unfiltered) fluorescence imageFigure C.1: False-color fluorescence images processed with a rolling-ball filter ofvarious ball sizes. False-color scale is the same for all images, and all images are150 x 150 ?m.198


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items