The discovery of monolayer graphene by Manchester group has led to intensive research into a variety of applications across different disciplines. As a monolayer of carbon atoms, graphene presents a high surface to volume ratio and a good electronic conductivity, making it sensitive to its surface bio-chemical environment. This project investigated the fabrication of electronic biosensors using different graphene-based materials. It included the production of graphene, the fabrication of electronic devices, the chemical functionalisation of graphene surface and the specific detection of target bio-molecules. This project first investigated the production of graphene using three different methods, namely mechanical exfoliation, physical vapour deposition and electrochemical reduction of graphene oxide. With respect to the physical vapour deposition method, the production of large area transfer-free graphene from sputtered carbon and metal layers on SiO2 substrate has, for the first time, been achieved. The relationship between growth parameters and the quality of resultant graphene layer has been systematically studied. In addition, a growth model based on the detailed analysis of morphological structures and properties of graphene film was simultaneously proposed. Optical microscopy, Raman spectroscopy and atomic force microscopy were used for the evaluation of the number, the quality and the morphology of resultant graphene layers in each method. To investigate the performance of graphene electronic devices, field effect transistors were fabricated using both exfoliated and chemical vapour deposited graphene. A novel technique for graphene patterning has been developed using deep ultraviolet baking and an improved photolithography method. A new shielding technique for the low damage deposition of Au electrodes on graphene has also been developed in this project. The practical challenges of device fabrication and performance optimisation, such as polymer residue and contact formation, have been studied using Raman spectroscopy and the Keithley 2602A multichannel source meter. For the functionalisation of graphene, a number of chemicals were investigated to provide linking groups that enable binding of bio-probes on the graphene surface. Hydrogen peroxide and potassium permanganate have been demonstrated to have the capability of immobilising oxygen-containing groups onto graphene. The levels of oxidation were estimated by energy dispersive analysis and Fourier transform infrared spectroscopy. In addition, aminopropyltriethoxysilanes and polyallylamine have exhibited good efficiency for immobilising amino groups onto graphene. The resultant graphene was characterised by X-ray photoelectron spectroscopy and cyclic voltammetry measurements. Graphene electrodes modified with electrochemically reduced graphene oxide were developed for the first time which exhibit significantly improved redox currents in electrochemical measurements. Using single stranded DNA immobilised via π-π bonds as probes, these electrodes showed a limit of detection of 1.58 x 10-13 M for the human immunodeficiency virus 1 gene. In parallel, human chorionic gonadotropin sensors were developed by immobilising its antibodies on 1-pyrenebutyric acid N-hydroxysuccinimide ester functionalised graphene field effect transistors. These field effect transistors have been demonstrated to exhibit a quantitative response toward the detection of 0.625 ng/ml antigen. In summary, the fabrications of two types of graphene-based biosensors for the detection of specific DNA sequence and human chorionic gonadotropin have been achieved in this project. Their sensitivity, selectivity, reproducibility and capability of multiple biomarker detection need to be further improved and explored in future work. The outcomes of this project have provided not only ready-made biosensing platforms for the detection beyond these two targets, but also novel techniques applicable to the development of multidisciplinary applications beyond biosensor itself.

Mercury (Hg) is a global pollutant and potent neurotoxin that is detrimental to the environment and human health. (MeHg). All forms of Hg are toxic, but methylmercury (MeHg) can biomagnify through food webs and become concentrated in food staples such as fish and rice, creating an exposure risk to people. The conversion of Hg to MeHg is mediated by anaerobic microbes, particularly sulfate and iron-reducing bacteria and methanogenic archaea. However, Hg methylation is an intracellular process, and MeHg production is dependent on the bioavailability of inorganic Hg to these microbes. One outstanding knowledge gap in understanding Hg methylation is the nature of bioavailability of inorganic divalent Hg (HgII). Much research has gone into developing a framework describing how microbes take up Hg for methylation. Still, the framework describing Hg bioavailability processes is not fully developed. The overall objective of my thesis is to address these mechanisms governing HgII bioavailability to anaerobic microbes. HgII bioavailability is determined by its speciation; these are all the different forms and compounds of HgII. To address the bioavailability of various HgII species, I used microbial Hg-biosensors. Hg-biosensors are bacterial cells that emit a quantifiable signal when HgII enters them and let me observe HgII bioavailability in real-time. The biosensors I developed are the first Hg-biosensors that function without oxygen and let me explore HgII species and their bioavailability under conditions conducive to methylation. HgII speciation is spatially and temporally dynamic moving from oxic to anoxic conditions and under various biogeochemical controls. I follow HgII speciation and bioavailability in my thesis as it transgresses through these conditions. Understanding HgII bioavailability to complex microbial communities across redox gradients and through dynamic ligand interactions is a missing key component to understanding and predicting MeHg formation. My results show how altering HgII speciation can identify novel bioavailability pathways or make it completely inaccessible. My results highlight how microbes can control HgII bioavailability and the importance of microbial community structure on metal acquisition. First, I resolve pathways for charged inorganic HgII species through the cell membrane and demonstrate novel pathways for previously unconsidered charged species. Using dissolved organic matter (DOM) originating from various algal species, I show how algae can uniquely control HgII bioavailability to other organisms. I demonstrate how DOM has emergent properties that can control HgII bioavailability. Next, I investigated the compounds microbes use to scavenge metals such as iron and copper. I reveal how they could inadvertently interact with HgII and form new bioavailability pathways. Lastly, I demonstrate how the diffusion of biogenic hydrogen sulfide from an isolated system can make an otherwise non-bioavailable HgII species rapidly available for microbial uptake. Overall, my thesis expands the framework describing HgII bioavailability to microbes and potential drivers of Hg methylation in the environment.

Plant mechanisms regulating environmental heavy metal interactions are vital for plant survival. Plants must maintain adequate metal levels while preventing excesses. Several mechanisms involved in heavy metal uptake and sequestration have been identified and studied in hyperaccumulating plants such as Thlaspi caerulescens. These plants accumulate large quantities of metals, but their use in remediation is limited by their small size. On the other hand, mechanisms in high-biomass, non-hyperaccumulating perennial species such as poplar (Populus spp.) are unknown. The central goal of this project is to delineate specific mechanisms in poplar regulating the heavy metal zinc (Zn) for potential use in bioremediation and real-time monitoring. Specifically, project aims are: 1) Determine the role of HMA4 and PCS1 genes in poplar; 2) Delineate the ZIP gene family including ZIP1 and ZNT1 activity; and 3) Harness fluorescent energy transfer to engineer a poplar tree that monitors Zn-soil contamination. These are addressed using current technologies including phylogenetic analysis, gene transformation, expression assays, promoter-GUS assays, fluorescent-gene imaging, and metal assays. Through these experiments, mechanisms controlling heavy metal interaction are identified and characterized in poplar. Poplar contains a large number of genes in both the ZIP and HMA4 families, but only two members in the PCS family. Poplar also contains several genes that share close sequence and structural homology to those in hyperaccumulators. However, there is an overall divergence from hyperaccumulators in regards to expression across an environmental Zn gradient. Poplar tightly regulates Zn intake by suppressing absorption avenues under Zn excess. Over-expression of HMA4 and PCS1 resulted in more tolerance and more accumulation, respectively, in poplar lines. These findings support a regulatory system used in poplar to limit Zn under excess and promote Zn under deficiency. Using ZNT1 and its natural expression gradient, a chimeric protein was created that served as a biosensor in both poplar and Arabidopsis thaliana host plants and was able to discriminate between 1μM and 10mM Zn concentrations. These findings add to current knowledge of heavy metal regulation and help fill the gap of knowledge currently existing on the regulatory mechanisms that perennial trees use to control heavy metals.

The design of pathogen biosensors may soon incorporate beads having a nanoscale diameter, thus making the drag force on a nanoscale sphere an important engineering problem. Flows at this small of a scale begin to appear "grainy" and may not always behave as a continuous fluid. Molecular dynamics provides an approach to determine drag forces in those nanoscale flows which cannot be described with continuum (Navier-Stokes) theory. This thesis uses a molecular dynamics approach to find the drag forces acting on a sphere and a wall under several different conditions. The results are compared with approximations from a Navier-Stokes treatment and found to be within an order of magnitude despite the uncertainties involved in both the atomic interactions of the molecular dynamics simulation and the appropriate boundary conditions in the Navier-Stokes solution. / Master of Science

Three major topics will be discussed in this dissertation. The first is an optical biosensor for specific diagnosis of bacterial skin and wound infection, followed by a paper microfluidic assay and accompanying monitoring device for monitoring blood coagulation and determining patient-specific heparin and protamine dosing. The final work to be discussed is ongoing work involving the detection of circulating tumor cells (CTCs) using a paper microfluidic detection platform. All of these works involve the development of biosensors for the simultaneous advancement and simplification of diagnosis and analysis of blood and bacterial infection. The aims of each of these projects included significantly decreasing the time to diagnosis and decreasing the reagents, laboratory space, personnel, and other resources needed for detection and diagnosis. The first works are focused on the design, development, and testing of an optical biosensor for the immediate detection of bacterial skin and wound infection, including diagnosing the specific species of bacteria responsible for the infection. The optical biosensor developed allows for diagnosis of a bacterial infection on skin or in a wound in as little as three seconds, in a contact-free, reagent-free manner. The second work focused on the design, development, and testing of a paper microfluidic assay and accompanying Raspberry Pi-based monitoring device for use before, during, and after surgeries requiring the use of cardiopulmonary bypass. The assay monitors the extent of blood coagulation of a whole blood sample and determines patient-specific dose response curves of an anticoagulant and its reversal agent. The final work discussed focuses on developing a paper microfluidic assay for the detection of CTCs from whole blood samples. The goal of this work is to detect multiple morphologies of CTCs from whole blood samples to provide insight on patient prognosis in a rapid, low resource manner.

Recent advances in the development of immunoassays and nucleic acid assays have improved the performance and increased the sensitivity of sensors that are based on biochemical recognition. The new approaches taken by researchers include detecting pathogens by detecting their nucleic acids, using new nontoxic reporter entities for generating signals, and downscaling and miniaturizing sensors to micromigration and microfluidic formats. This dissertation connects some of these successful approaches, thereby leading to the development of novel nucleic acid sensors for rapid and easy detection of pathogens. The author's goal was to develop diagnostic tools that enable investigators to detect pathogens rapidly and on site. While the sensors can be used to detect any pathogen, the author first customized them for detecting particularly Cryptosporidium parvum, a pathogen whose detection is important, yet presents many challenges. Chapter 2 of this thesis presents a novel test-strip for the detection of C. parvum. The test-strip is designed to detect nucleic acids rather than proteins or other epitopes. While test strips are commonly used for sensors based on immunological recognition, this format is very new in applications in which nucleic acids are detected. Further, to indicate the presence or absence of a specific target on the test strip, dye-entrapped, oligonucleotide-tagged liposomes are employed. Using liposomes as reporter particles has advantages over using other reporter labels, because the cavity that the phospholipidic membranes of the liposomes form can be filled with up to 106 dye molecules. By using heterobifunctional linkers liposomes can be tagged with oligonucleotides, thereby enabling their use in nucleic acid hybridization assays. The developed test-strip provides an internal control. The limit of detection is 2.7 fmol/mL with a sample volume of 30 mL. In chapter 3 the detection of nucleic acids by means of oligonucleotide-tagged liposomes is scaled down to a microfluidic assay format. Because the application of biosensors to microfluidic formats is very new in the field of analytical chemistry, the first part of this chapter is devoted to developing the design and the method to fabricate the microchip devices. The performance of the microchips is then optimized by investigating the interactions of nucleic acids and liposomes with the material the chips consist of and by passivating the surface of the chips with blocking reagents. The developed microfluidic chip enabled us to reduce the sample volume needed for one assay to 12.5 mL. The limit of detection of this assay was determined to be 0.4 fmol/mL. Chapters 4 and 5 expand on the development of the microfluidic assay. A prototype microfluidic array that is able to detect multiple analytes in a single sample simultaneously is developed. Using such an array will enable investigators to detect pathogens that occur in the same environment, for example, C. parvum and Giardia duodenalis by conducting a single test. The array's ability to perform multiple sample analysis is shown by detecting different concentrations of target nucleic acids. Further, the author developed a microfluidic chip in which interdigitated microelectrode arrays (IDAs) that consist of closely spaced microelectrodes are integrated. The IDAs facilitate electrochemical detection of cryptosporidial RNA. Electrochemical detection schemes offer benefits of technical simplicity, speed, and sensitivity. In this project liposomes are filled with electrochemically active molecules and are then utilized to generate electrochemical signals. Chapter 6 explores the feasibility of liposomes for enhancing signals derived from nucleic acid hybridization in surface plasmon resonance (SPR) spectroscopy. SPR spectroscopy offers advantages because nucleic acid hybridization can be monitored in real time and under homogeneous conditions because no washing steps are required. SPR spectroscopy is very sensitive and it can be expected that, in the future, SPR will be integrated into microfluidic nucleic acid sensors. / Jüngste Fortschritte in der Entwicklung von Immuno- und Nucleinsäure- Assays haben die Arbeitsleistung und die Spezifität von Sensoren, die auf biochemischer Erkennung basieren (Biosensoren), verbessert. Neu entwickelte Methoden umfassen die Detektion von Pathogenen durch die Detektion ihrer RNA oder DNA, das Benutzen von neuen nicht-toxischen Reporter Molekülen, um Signale in Sensoren zu erzeugen, und die Verkleinerung und Miniaturisierung von Sensoren zu Mikromigrations- und Mikrofluid Formaten. Die in dieser Dissertation entwickelten Sensoren, die der Detektion von Pathogenen dienen, verbinden einige der neu entwickelten Methoden. Das Ziel der Autorin war es, Sensoren zu entwickeln, die es ermöglichen, Pathogene an Ort und Stelle zu detektieren. Die entwickelten Sensoren können zur Detektion von einer Reihe von Pathogenen benutzt werden. In dieser Dissertation sind sie für die spezifische Detektion von Cryptosporidium parvum entwickelte worden. Kapitel 2 der Dissertation präsentiert einen neuen Teststreifen für die Detektion von C. parvum. Der Teststreifen detektiert die RNA von C. parvum, die als Reaktion auf einen Hitzeschock produziert wird. Das Teststreifen-Format ist üblich für Sensoren, die auf immunologischer Erkennung basieren. Es ist jedoch neu für Anwendungen in denen RNA oder DNA detektiert werden sollen. Die An- oder Abwesenheit eines bestimmten Ziel Moleküls wird durch Liposomen, die Oligonukleotide auf der Aussenseite ihrer Membranen enthalten und mit Farbstoff gefüllt sind, angedeutet. Die Experimente zeigten, dass die mit dem entwickelten Test-Streifen kleinste detektierbare Konzentration von RNA in einem 30 mL Probenvolumen 2.7 fmol/mL ist. In Kapitel 3 ist die Signalerzeugung durch Liposomen in ein Mikrofliess-System integriert. Da die Entwicklung von Mikrofliess-Systemen ein sehr neues Forschungsgebiet ist, befasst sich ein Teil dieses Kapitels mit dem Design und der Herstellung des Microchips. Die Untersuchung von Interaktionen von Nukleinsäuren und Liposomen mit dem Material aus dem der Chip hergestellt ist und die Passivierung dieses Materials ist dabei ein Schwerpunkt. Das Probenvolumen, dass zur Detektion mit dem entwickelten Mikrofliess-Sensor nötig ist, konnte auf 12.5 mL reduziert werden. Die kleinste detektierbare Konzentration von Nucleinsäuren ist 5 fmol/mL. In Kapitel 4 und 5 erweitert die Autorin die Entwicklung des Mikrofliess-Sensors aus Kapitel 3. Das Detektionsformat ist auf ein Array, das für die gleichzeitige Detektion von mehreren Pathogenen benutzt werden kann, angewandt. Eine Methode zum Herstellen eines Arrays-Prototypen ist entwickelt. Ferner, stellte die Autorin verzahnte Mikroelektroden her und benutzte diese um die elektrochemische Detektion der RNA von C. parvum zu ermöglichen. In Kapitel 6 ist die Anwendbarkeit von Liposomen zur Erhöhung von Signalen von Nukleinsäure-Hybridisierungen in Surface Plasmon Resonance Spectroscopy (SPR) untersucht.

Thesis advisor: Thomas Chiles / Molecular recognition is an important area of research as it has far reaching applications in sensors, molecular separations, and medicine. Molecularly imprinted polymers offer an option for developing high resolution tools of detection that are both selective and sensitive. As a platform, carbon nanotubes offer a highly conductive surface and their growth and unique magnetic properties can be manipulated for our purposes. Such carbon-nanotube based sensors can afford high sensitivity, while molecular imprinting provides the selectivity of detection with the flexibility of fabrication. In order to fabricate a molecular imprint, monomeric compounds are polymerized in the presence of a target molecule of interest, which acts as the template. Once the template molecule has been removed an imprint capable of “recapturing” the target molecule is left behind. In this work we used cyclic voltammetry as a means of depositing polymer coatings doped with a target molecule. We fabricated a molecularly imprinted polymer sensor specific for ferritin using polyphenol as the polymer. The development of our imprint was monitored based on changes in impedance levels calculated by electrochemical impedance spectroscopy. After depositing ferritin-doped polyphenol layers we evaluated the effectiveness of different eluant solutions. Ultimately, deionized water was determined to be the developing solution of choice because it effectively removed the ferritin while not compromising the integrity of the remaining polymer coating. The sensor was capable of detecting ferritin at a concentration of 1x10-9 g/L (1 pg/mL). In parallel we evaluated the stability of the polyphenol coating. / Thesis (BS) — Boston College, 2009. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: College Honors Program. / Discipline: Biology.

Microfluidic biosensors and bioreactors based on immobilized biomaterials are described in this dissertation. Photocrosslinkable hydrogel or polymeric microbeads were used as a supporting matrix for immobilizing E.coli or enzymes in a microfluidic device. This dissertation covers a microfluidic bioreactor based on hydrogel-entrapped E.coli, a microfluidic biosensor based on an array of hydrogel-entrapped enzymes, and a microfluidic bioreactor based on microbead-immobilized enzymes. Hydrogel micropatches containing E.coli were fabricated within a microfluidic channel by in-situ photopolymerization. The cells were viable in the hydrogel micropatch and their membranes could be porated by lysating agents. Entrapment of viable cells within hydrogels, followed by lysis, could provide a convenient means for preparing biocatalysts without the need for enzyme extraction and purification. Our results suggested that hydrogel-entrapped cells, immobilized within microfluidic channels, can act as sensors for small molecules and as bioreactors for carrying out reactions. A microfluidic biosensor based on an array of hydrogel-entrapped enzymes could be used to simultaneously detect different concentrations of the same analyte or multiple analyte in real time. The concentration of an enzyme inhibitor could be quantified using the same basic approach. Isolations of the microchannels within different microfluidic channels could eliminate the possibility of cross talk between enzymes. Finally, we characterized microfluidic bioreactors packed with microbead-immobilized enzymes that can carry out sequential, two-step enzyme-catalyzed reactions under flow conditions. The overall efficiency of the reactors depended on the spatial relationship of the two enzymes immobilized on the beads. Digital simulations confirmed the experimental results.

Pressure ischemia is a medical condition characterised by the necrosis of the skin and underlying tissues in body areas exposed to prolonged pressure. This condition leads to the development of bedsores and affects 9% of hospitalised patients, costing the NHS between £1.4 and £2.1 billion per year. The severity of pressure ischemia has been linked to the concentration of sweat lactate, a product of sweat gland metabolism under anaerobic conditions, such as hypoxia. Normal levels of lactate in human sweat are 20±7 mM, but under ischemic conditions these can rise up to approximately 70 mM. This project presents the development of a novel flexible electrochemical enzyme-based biosensor for the continuous and non-invasive monitoring of sweat lactate with the potential for becoming a body-worn device for the early detection of pressure ischemia onset. The core of the recognition system is a flexible laminate, comprising two highly porous polycarbonate membranes, which provide support for the lactate oxidase enzyme, immobilised via covalent cross-linking. Oxidation of lactate produces H2O2, which is subsequently determined electrochemically. The transducer comprises a two-electrode system on a single flexible polycarbonate membrane, sputter-coated with gold (CE/RE) and platinum (WE) to render it conductive. The developed design has been improved through investigation into different factors regarding the immobilisation method of the enzyme in the laminate and the lowering of interferences from oxidising compounds present in sweat. The sensing system exhibits lactate selectivity at physiologically relevant concentrations in sweat for pressure ischemia (0–70 mM), with good reproducibility (7.2–12.2% RSD) for a hand-manufactured device. The reliability of the sensor’s performance and the capability to detect lactate fluctuations on human sweat samples has been demonstrated. The sensing system showed excellent operational and mechanical stability. The application of Nafion® on the WE lowered interferences from ascorbic acid and uric acid by 96.7 and 81.7% respectively. These results show promise towards the further development of a body-­‐worn monitoring device for determining lactate levels in undiluted human sweat samples in a reproducible, fast and accurate manner.

Zugl.: Mainz, Univ., Diss., 2005