Semiconductor nanowires (NWs) are one- or quasi one-dimensional systems whose physical properties are unique as compared to bulk materials because of their nanoscaled sizes. They bring together quantum world and semiconductor devices. NWs-based technologies may achieve an impact comparable to that of current microelectronic devices if new challenges will be faced. This thesis primarily focuses on two different, cutting-edge aspects of research over semiconductor NW arrays as pivotal components of NW-based devices. The first part deals with the characterization of electrically active defects in NWs. It has been elaborated the set-up of a general procedure which enables to employ Deep Level Transient Spectroscopy (DLTS) to probe NW arrays’ defects. This procedure has been applied to perform the characterization of a specific system, i.e. Reactive Ion Etched (RIE) silicon NW arrays-based Schottky barrier diodes. This study has allowed to shed light over how and if growth conditions introduce defects in RIE processed silicon NWs. The second part of this thesis concerns the bowing induced by electron beam and the subsequent clustering of gallium arsenide NWs. After a justified rejection of the mechanisms previously reported in literature, an original interpretation of the electron beam induced bending has been illustrated. Moreover, this thesis has successfully interpreted the formation of NW clusters in the framework of the lateral collapse of fibrillar structures. These latter are both idealized models and actual artificial structures used to study and to mimic the adhesion properties of natural surfaces in lizards and insects (Gecko effect). Our conclusion are that mechanical and surface properties of the NWs, together with the geometry of the NW arrays, play a key role in their post-growth alignment. The same parameters open, then, to the benign possibility of locally engineering NW arrays in micro- and macro-templates.

This study is a natural follow on from previous work by M. T. Anderson and I. Fiebrig. The goal of those latter and of the present study is to find a mucoadhesive system for improving the oral bioavailability of a number of drugs, for example bioactive peptides and proteins. This current work evaluates the adhesive properties of a cationic polymer and a cationic protein to mucus glycoproteins as a step towards the future development of a mucoadhesive drug delivery system. Four different mucin populations were analysed in solution (a freshly purified sample PGM-MD, and three purified from different regions of the porcine stomach cardiac, antrum and fundus). Their interaction with two groups of chitosans differing in degree of deacetylation (FA = 0.11 and 0.25) and a protein purified from the foot of the blue mussel Mytilus edulis foot protein-1 (Mefp-1) were studied. Interaction was determined using analytical ultracentrifugation and with the chitosan/mucin interaction specifically atomic force microscopy. The influence of ionic strength on the interaction was studied in detail studied as was the effect of the oligosaccharide composition of the mucin population on the interaction. It was found that both groups of chitosans (FA = 0.11 and 0.25) formed a large complex with a freshly purified mucin population (PGM-MD). Ionic strengths above 0.2 M were found to inhibit the interaction. The three mucin species differed in terms of their net charge, with cardiac being the most negatively charged and antrum the least negative. It was found that the cardiac species interacted the most and antrum the least, as would be expected for an ionic interaction. Increasing ionic strength was found to inhibit the interaction. There was also evidence for a hydrophobic interaction at high ionic strengths. The atomic force microscopy results allowed the complex to be visualised under atmospheric conditions and to get away from the harsh sample preparation techniques employed by electron microscopy. Large spherical complexes were seen as entanglements of mucin and chitosan strands.

The weight variation of gelatin capsules filled by a dosator-type machine has several possible causes, one of which, the presence of density variations within the powder feed bed, has been evaluated, using a number of particle size fractions of material. Existing theories suggested that the bulk density, or porosity of a powder bed depends on the velocity and intensity of deposition of the particles, together with the properties of the powder. Experiments with lactose confirmed this, indicating in particular that low velocity deposition encourages the formation of regions of relatively high porosity. The application of vibration, especially in a vertical direction, proved effective in reducing the porosity of powder packings, and optimum frequency and acceleration appeared to be largely independent of particle size. A system was developed for detecting porosity variations within powder beds, using the technique of gamma-ray attenuation, the principle being that the reduction in intensity of a Gamma-ray beam traversing a powder bed is a function of local porosity. The observed linear attenuation coefficient of lactose was found to be a function of porosity, for reasons not fully established, hence a calibration expression for the attenuation coefficient was proposed, and was used in subsequent local porosity determinations. The radial distribution of porosity within cylindrical samples of lactose, prepared under various conditions, was studied, and a means of presenting such distributions pictorially was developed. The influence of deposition method, and applied vibration, on the uniformity, of such packings, was clearly demonstrated. A semi-automatic dosator-type capsule filling machine was used to evaluate the relationship between powder feed bed porosity variations and capsule fill weight variation. The results indicated that in the case of lactose, the influence of porosity variations, of the order of magnitude encountered, tends to be masked by other phenomena affecting the filling process.

Protein-based therapeutics play an increasingly important role in medicine, and the exquisite bio-recognition properties exhibited by antibodies has also led the their use in a number of other fields apart from medicine. The increasing use of these molecules requires more efficient methods of purification. A review of the current purification strategies was conducted. Of all the purification methods studied, peptide epitope/mimotope affinity chromatography proved to be the method of choice - resulting in antibodies exhibiting high specific immunoreactivity. The need to tailor unique affinity ligands for each antibody to be purified by peptide epitope/ mimotope affinity chromatography was identified as the major problem with this technique. A review of the technologies available to regulate the antibody-antigen interaction was conducted. Little was published on the use of phage display for the discovery of peptide ligands for use in peptide mimotope affinity chromatography. Experiments were conducted using a polyvalent phage display library to identify novel peptide ligands for the purification of the therapeutic monoclonal antibody C595. A novel peptide was isolated which demonstrated improved chromatographic performance compared to the standard epitope peptide used to purify mAb C595 from biological supernatants. Circular dichroism showed that the novel peptide had a more highly ordered structure at 4oC and room temperature than the epitope peptide, and fluorescence quenching revealed a higher equilibrium association constant. A method for the optimisation of peptide mimotopes derived from phage display, cross-reactive with an anti-steroid antibody was investigated. Improvements relating to the selection of lead peptide sequences are described, and the use of mimotopes in an assay to determine concentrations of steroid in solution has been demonstrated. The optimised mimotope was used as an effective paratope-specific affinity ligand. A novel method for selecting high affinity antibody fragments in vivo is described. The C595 scFv gene was fused to a gene encoding green fluorescent protein and incorporated into the phagemid vector pCANTAB 5E. The fusion protein could not be expressed at high levels and could only be detected using epitope affinity chromatography in combination with ELISA.

The chemistry and structure of Poly (glycerol adipate) facilitate its substitution with various pendant functional groups leading to modifications of the physicochemical properties of the polymer. Modified backbones then can be selected based upon the properties of the compound to be incorporated. Thus, this could be explored as a drug delivery system without many of the limitations of commercially available polymers. The aim of this study was investigate whether various polymers and drugs interact in a specific manner and whether the nature of these interactions influence the physicochemical characteristics of the particles and their drug loading and release profile. By investigating drugs belonging to various classes and with different properties it has been possible to correlate properties associated with drugs and pendant functional groups of the polymer which are ultimately responsible for the drug loading and release characteristics. For some drug polymer formulations, good loading and controlled release rates have been achieved. Compared to various conventional polymer systems reported for nanoparticle formulations, poly (glycerol adipate) polymers have also demonstrated the ability to control rate of release of highly water soluble drugs, even from the most hydrophilic polymer backbone in its unsubstituted form. From the various drug loading and release profiles it has been demonstrated that, unlike reported literature, particle size is not the primary factor influencing drug release over the relatively small range of particle sizes seen in this study. Neither is the water solubility of either the drug or the polymer alone responsible for the rapid and uncontrolled release profile from nanoparticles. Thus, Drug polymer interactions are more likely to influence drug loading and release and unlike common reports in the literature, hydrophilicity, molecular weight or concentration of polymer / drug are less likely to affect these parameters in isolation.

Some of the colloidal aspects of perfluorochemical emulsions have been investigated. Particular attention was given to the influence of the nature of the oil phase on emulsion stability. Bulk emulsion stability was measured by an electron micrographic technique. Interfacial and single droplet rest-time data were also collected. A range of surfactants and perfluorochemicals were investigated. It was found that emulsion stability depends on the chemical nature of the oil phase and the emulsifier. The differences in stability could be rationalized in terms of the intermolecular forces between oil molecules, and oil and surfactant molecules. The effect on stability of a small amount of an additive incorporated into the oil phase was also investigated. It is postulated that although coalescence is the main mechanism by which fluorocarbon emulsions coarsen, molecular diffusion (Ostwald Ripening), in the more stable systems, is also important. Most stable emulsions were obtained by utilizing an emulsifier system comprising a small and a large molecular weight emulsifier. Accelerated stability testing and optimum storage conditions were also investigated. About 40C was found to be the optimum storage temperature. The problem of the fluoride ion production during emulsification could be minimised by sonicating in a carbon dioxide atmosphere. The oxygen uptake and release by fluorocarbon emulsions was rapid, reaching equilibrium within half a second. The in-vitro phagocytosis experiments showed that the phagocytosis rate of fluorocarbon emulsions was dependent on the droplet diameter and its surface characteristics. Investigation of methods to sterilize the fluorocarbon emulsions showed that filtration of constituents before emulsification coupled with autoclaving had the minimum effect on stability. A qualitative correlation between single droplet stability and bulk emulsion stability was found and it is concluded that the method could be a useful screening procedure to find an optimum system.

This thesis describes the results from a series of studies designed to evaluate the gastrointestinal transit of oral dosage forms. The transit of placebo pellet and tablet formulations was monitored using the technique of gamma scintigraphy. The formulations were radiolabelled with either technetium-99m or indium-lil. Four parameters, two physiological and two pharmaceutical, were selected for investigation. All the studies were conducted in healthy male volunteers. The first study examined the influence of the supine position on the gastric emptying of pellets in fasted and fed subjects. There was no marked difference between the supine and control gastric emptying data. As would be expected, food had a significant effect on gastric emptying. The influence of the time of day of administration on the gastrointestinal transit of pellets was investigated in fasted subjects. Transit of the pellets was not affected by their time of administration. The effect of the putative bioadhesive, polycarbophil, on the gastrointestinal transit of a pellet formulation was studied in fasted subjects. The pellets emptied from the stomach, rapidly and in an exponential manner. A set of studies was conducted to evaluate the transit of tablets in fed and fasted subjects. Tablet size did not affect gastric emptying, although there was an increase in the variability of gastric emptying with increasing tablet size. Food had a marked effect on gastric emptying. The rate of emptying was related to the energy content of the meal. Tablet size did not appear to be a determinant of transit through the ileocaecal sphincter. The colon transit and dispersion of the tablets was examined. Neither the ingestion of food nor defecation appeared to alter the rate of transit through the colon.

Hard gelatin capsules are manufactured from blends of limed ossein (LOG), acid ossein and acid pigskin gelatins, with LOGS usually forming the main portion of the blend. Unfortunately, the quality of the finished capsule cannot be predicted by routine quality control tests and 'poor' batches of LOG are encountered which form unsatisfactory capsules. In addition, differences between gelatin types are encountered in practice but the supporting literature is sparse and conflicting. This work examines some properties of gelatins pertinent to hard capsule manufacture in an attempt to relate these to gelatin type and known performance on capsule machines. Molecular mass distributions (MMD) were examined by polyacrylamide gel electrophoresis. Each different gelatin type possessed a characteristic shape of MMD which could be related to the source tissue and the treatment undergone during gelatin manufacture. MMDs, isoionic point determinations and an assessment of the content of protein impurities by ultraviolet spectrophotometry failed to resolve differences between 'good' and 'poor' LOG batches. The drying rates of freshly-cast gelatin films were studied under conditions of controlled humidity, temperature and air velocity. No significant differences between gelatin types were observed. Small variations in equilibrium moisture content were seen and were tentatively ascribed to the generation of water binding sites (free carboxyl groups) during gelatin manufacture. The mechanical properties of dry gelatin films were examined by tensile stress-strain and stress-relaxation measurements. Viscoelastic behaviour typical of a polymer in the glass transition region was observed, and significant differences in film fracture strain were observed but these were not related to gelatin type. Optical rotation measurements indicated similar orders of film crystallinity and determinations of frictional characteristics similarly revealed no varietal differences. Overall, few differences related to gelatin type were seen within the properties examined. There was no evidence in these studies to explain the unsatisfactory behaviour of 'poor' LOG batches in capsule manufacture. NB. This ethesis has been created by scanning the typescript original and may contain inaccuracies. In case of difficulty, please refer to the original text.

The aim of this project is to compare pharmaceutical particles made using a Nektar supercritical fluid technology technique called solution enhanced dispersion by supercritical fluids (SEDSTM) to those made using more traditional techniques. This involves a comparison of not only the surface properties of both types of particles, but also the interparticulate interactions. The majority of the work has involved the use of the atomic force microscope (AFM) as both a tool for imaging and for the acquisition of localized force measurements. The first experimental chapter of this work describes a method developed in order to image the contacting asperities of a particle. The AFM has the potential to provide useful information regarding single particle interactions to complement data generated from bulk techniques. In this chapter, the AFM artefact of tip imaging was used to produce 3D images of the asperities of particles of micronised and SEDSTM salbutamol sulphate, an anti-asthma drug, contacting a model surface of highly orientated pyrolytic graphite (HOPG). These data were recorded in a model propellant environment, used in order to simulate the environment that would be found in pressurised metered dose inhalers, such as those used by asthmatics. From the images generated the contacting area was estimated to be 1.1x10-3 mm2 for the micronised material, and 1.4x10-3 for the SEDSTM material. The work of adhesion for both of the materials was also calculated, and the values of 19.0 mJm-2 and 4.0 mJm-2 were obtained for the micronised and SEDSTM samples respectively. This supported available data that indicated the SEDS material had a lower surface energy than the micronised drug, and that it is possible to make comparisons between different modified AFM probes. The second chapter develops this work so that it can be applied to an air environment, which is applicable to more pharmaceutical systems. Here, force measurements were again performed using AFM, with the same drug samples studied in the first chapter, except a controlled relative humidity (RH) environment was used, so that the variation in adhesion with increasing RH could be studied. Two types of measurement were undertaken. The first involved the use of blank AFM tips on compressed disks of drug material, and the second involved the use of drug particles mounted onto AFM tips on both HOPG and compressed disks of drug. With the blank AFM tip and particle modified AFM tip on HOPG work it was observed that the SEDSTM materials showed a peak in adhesion force at 22% RH while the micronised salbutamol showed a peak at 44% RH. From this, a three-scenario model of linking morphology of contact to adhesion was developed to explain the observed peaks in adhesion. In addition, the surface energies of each of the two samples were calculated using the force measurements acquired against HOPG and compressed disks of material and compared. The micronised material was found to have a higher surface energy than the SEDSTM material (10.8 mJm-2 cf 5 mJm-2) when data acquired against HOPG was used. However, when data acquired using the compressed disks of drug were used, the SEDSTM had a higher surface energy than the micronised (29.9 mJm-2 cf 22.6 mJm-2). This higher value was attributed to different surface roughness effects found with the compressed disks. The third chapter uses the techniques and models developed in the previous chapters to examine the effect of polymorphism on surface energy, structure and particulate interactions. Three polymorphs of the drug sulphathiazole (forms I, II and IV) were formed using the SEDSTM technique, one of which (form I) was formed using two different solvents: methanol and acetone. Force measurements were performed using the AFM at controlled humidity using particles of each of the polymorphs mounted onto AFM tips against substrates of HOPG and the polymorph under analysis. This data was then related to the model developed in the previous chapter, and calculations were undertaken to assess the different surface energies of each of the four samples. For some of the samples it was observed that peaks were again occurring in the data, at 22% RH for polymorphs I-methanol and III, and 44% for polymorph IV. No peak was seen for polymorph I-acetone. These peaks were then related to the surface energy calculated for each of the polymorphs, as polymorphs I-methanol and III were found to have lower surface energy (0.99 mJm-2 and 1.17 mJm-2 respectively) than polymorphs IV and I-acetone (20.33 mJm-2 and 309 mJm-2). The fourth chapter examines the application of AFM to an industrial problem. When using the SEDSTM process to manufacture insulin, it was observed that the SEDSTM material had poorer flow properties than that of the unprocessed material. Using the AFM as both an imaging and force measurement tool, this chapter explores the application of imaging and the adhesion models and surface energy calculations previously developed to understand this problem. The AFM images showed the presence of highly aggregated particles of SEDSTM insulin, compared to the unprocessed insulin that appeared to be more crystalline. When force measurements were performed against both HOPG and particles of the material under analysis, non of the unprocessed, and only one of the SEDSTM particle tips prepared displayed the peak behaviour seen with previous measurements, and instead displayed a continual increase in adhesion force with humidity. In addition, when the surface energy was calculated, the SEDSTM material was found to be higher than the unprocessed insulin (77.5 mJm-2 cf 2.4 mJm-2). The increase in adhesion force was related to the particles agglomerating together, due to the presence of a higher surface energy and high amorphous content of the particles. The final experimental chapter uses techniques that compliment AFM analysis to examine another industrial problem. The SEDSTM process can be used to co-formulate drugs with other materials such as polymers. In this chapter, the drug pregabalin has been co-formulated with lipid in order to produce a coating around the drug to mask taste. The use of AFM as an imaging tool, and the additional techniques of X-ray photoelectron spectrometry (XPS) and time-of-flight secondary ion mass spectroscopy (ToF-SIMS) have been used to generate an understanding of surface structure and chemistry of this heterogeneous system. The AFM images showed no areas of surface heterogeneous behaviour, although the largest scan size was only 5 mm x 5 mm. However both the XPS and ToF-SIMS spectra, which samples far larger areas (up to 75 mm x 75 mm) showed the presence of lipid and drug molecules. It was concluded that the lipid was not forming a uniform layer around the drug molecule, but was instead forming large patches that were beyond the resolution of the AFM. This work aims therefore to provide a fundamental study of the application of AFM to real pharmaceutical systems. In particular models are developed which allow not only ranking of particle interactions but the quantification of factors such as surface energy and work of adhesion. Finally the significance of the morphology of the inter-particulate contact has been explored at the nanoscale.

It has been widely reported that cell culture dimension and microenvironment influence cell proliferation, differentiation, and gene expression, which lead to different interactions between drug delivery systems and cells. The development in evaluation of drug delivery systems has reached the stage where investigations are now concentrating on intracellular uptake and subcellular localization of drug delivery systems.This thesis investigates the use of three-dimensional (3-D) tissue culture models to study how nanoparticles (NPs) may behave in vivo. Poly (glycerol-adipate) (PGA) NPs can degrade into glycerol and adipate, which are not having toxic and anyundesirable local or systemic effects in the host. Following on the initial physicochemical characterization of PGA NPs loaded with drug and fluorescent dyes, investigations moved on to the biological studies of NPs in various cell culture model, e.g. monolayer culture, 3-D culture models, and brain tumour invasion model. Particle size, surface charge, and hydrophobicity are important features affecting the amount of particles taken up by cells and intracellular localisation of particles. Thus, the physicochemical properties of drug and fluorescent dye loaded PGA NPs were assessed by Photon Correlation Spectroscopy, Laser Doppler Anemometry, and drug/fluorescent dye loading studies. These studies indicated that physicochemical properties of drug, fluorescent dyes and PGA polymer could influence drug /fluorescent dye loading, which results in different particle size and surface charge of PGA NPs. Quantitative and qualitative investigations into the influence of cell culture dimension on uptake of NPs by cells, both by confocal fluorescence microscopy and flow cytometry, revealed that DAOY cells took up NPs more effectively when in 3-D spherical aggregate culture than in 2-D monolayer culture while uptake of NPs by normal brain cells was lower in 3-D cell culture than that seen in 2-D monolayer culture. This resulted in intracellular fluorescence intensity about 6 times higher in DAOY aggregates than normal brain cell aggregates while in monolayer culture mixed brain cells took up 2 times as many NP as the DAOY cells. The results from studies of NPs migrating through aggregates and tissue slices also indicated that penetration ofNPs in 3-D culture models was affected by the structure of the interstitial compartment and composition of extracellular matrix. Microscopic investigation of the histology of a co-culture invasion model of DAOY aggregates and a organotypic brain slice confirmed that DAOY cells massively invaded into cerebellum slices after a 4-day co-culture while the invasion of DAOY cells were limited within cerebral cortex slices even after a 6-day co-culture. Selective uptake of NPs by host cells and brain tumour cells were also assessed in this 3-D brain tumour invasion model. It showed that most NPs were taken up by DAOY cells instead of brain cells.