Au cours des dernières années, la demande de plus en plus forte pour des mémoires non-volatiles performantes, a mené au développement des technologies NOR Flash et NAND Flash, qui dominent aujourd'hui le marché des mémoires non-volatiles. Cependant, la miniaturisation de ces technologies, qui permettait d'en réduire le coût, laisse aujourd'hui entrevoir ses limites. En conséquence, des mémoires alternatives et émergentes sont développées, et parmi celles-ci, la technologie des mémoires à changement de phase, ou PCRAM, est l'une des candidates les plus prometteuses tant pour remplacer les mémoires Flash, notamment de type NOR, que pour accéder à de nouveaux marchés tels que le marché des SCM. Toutefois, afin d'être pleinement compétitives avec les autres technologies mémoires, certaines performances de la technologie PCRAM doivent encore être améliorées. Au cours de cette thèse, nous cherchons donc à obtenir des dispositifs PCRAM plus performants. Parmi les résultats présentés, nous réduisons les courants de programmation et la consommation électrique des dispositifs tout en augmentant la rétention de l'information à haute température. Pour cela, nous modifions la structure du dispositif ou nous utilisons un matériau à changement de phase alternatif. De plus, à l'aide de solutions innovantes, nous permettons aux dispositifs PCRAM de conserver l'information pendant une éventuelle étape de soudure de la puce mémoire. Enfin, nous avons conçu, développé et validé un procédé de fabrication permettant d'intégrer une diode PN de sélection en Silicium en série avec un élément résistif PCRAM, démontrant l'intérêt de ce sélecteur vertical pour être utilisées comme élément de sélection d'une cellule PCRAM intégrée au sein d'une architecture crossbar. / In the past few years, the increasing demand for high quality non-volatile memory (NVM) devices, leads to the developpment of NOR Flash and NAND Flash technologies, which are now the two main NVM players. However, because of some limitations such as performance degradation and limited cost reduction, the scaling of these technologies will reach in the next few years. Therefore, new NVM technologies are under development and among them, phase-change memory (PCM) technology has attracted strong interest and is now became one of the most promising candidates in order to replace Flash technologies, especially NOR Flash technology, and to address new memory markets, such as storage-class-memory market. However, in order to fully take their role in the memory arena, some performances of the PCM technology have to be improved. Therefore, during this PhD, we have tried to improve PCM devices electrical performances by reducing both programming currents and energy consumption while increasing high-temperature data-retention. To this extent, we have studied innovative device structure and alternative phasechange material. Moreover, using innovatives solutions, we show that our PCM devices could store data during the soldering step of the memory chipset. Finally, we have conceived, developed and validated, a process flow in order to make 1D1R PCM cell with Silicon-based vertical PN diodes, proving the relevance of this selector for PCRAM-based crossbar architecture.

We characterize the impact of high-energy, 172 nm vacuum ultraviolet photons on the molecular weight and the glass transition temperature of poly(methyl methacrylate). We found that the molecular weight is reduced strongly on the surface of the exposed samples with a continuous transition towards the unexposed bulk material being located below the modified region. The glass transition temperature was found to be significantly lowered in the exposed region to well below 50°C compared to that of the 122°C of the bulk region. We could use this material contrast to selectively reflow the top surface of the exposed samples only. This allowed us to create ultra-smooth micro-optical structures by postprocessing without influencing the overall geometry that is required for the optical functionality.

In resting muscle, blood flow is regulated to meet the demand for O2 by the tissue. A modified ischemia (I)/reperfusion(R) investigation was systematically run and PISFO2, PaO2, Q and VO2 were observed. Twenty-nine spinotrapezius muscles from male Sprague-Dawley rats (284±20 grams) were surgically exteriorized for intravital microscopy to test a model relating blood flow, O2 supply and O2 demand. The model can aid in the understanding of the regulation of tissue PO2. The interstitial PO2 (PISFO2) and perivascular PO2 (PaO2) measurements were made using phosphorescence quenching microscopy (PQM). O2 consumption (VO2) values were obtained with a quasi-continuous, flash-synchronized, pressurized airbag to initiate ischemia and sample the rate of O¬2 change (dPO2/dt). Centerline red blood cell velocity was measured with an Optical Doppler Velocimeter and converted to flow using vessel diameter. 5-, 15-, 30-, 60-, 300- and 600-second ischemic durations were used to observe changes in PISFO2, Q, and VO2. A critical point was observed following 30 seconds of (I) where dPISFO2/dt during recovery was the fastest (4.25±0.72 mmHg/s) and was 1.00±0.16 mmHg/s following 600 seconds. Flow recovery, dQ/dt, peaked to 3.88±0.64 (µl•min-1)/s after 60 seconds of (I) but significantly dropped to 2.83±0.55 (µl•min-1)/s following 300 seconds of (I) but increased to 2.92±0.45 (µl•min-1)/s following 600 seconds. This gives evidence to a no-reflow phenomenon occurring in the extended periods of ischemia. A peak in VO¬2 to 309.2±45.0 nl O2/cm3•s with a time course of 160 seconds occurred following 600 seconds of ischemia. As the ischemic duration decreased, the time course and peak VO2 also decreased. VO2 following 300 seconds of (I) was significantly higher than 5-60 seconds of (I) (p <0.05) but was not significantly different from 600 seconds of (I). The information collected during the Q and VO2 studies can be incorporated into a factor, M, that relates VO2, Q and ∆PO2. M calculated for the recovery of 5- through 60-second (I) groups reasonably relates the three variables due to consistency and little variability. However, recovery in 600- and especially 300-second (I) groups showed higher variability in M which requires more consideration.

MMed thesis - Faculty of Health Sciences / The field of organ and tissue transplantation has necessitated an improved understanding of their associated pathophysiological pathways. Specific areas of interest involve the changes that follow ischaemia and derangement’s that accompany organ and tissue storage, reperfusion injury and the “no-reflow” phenomenon. Strategies have been devised to manipulate and modify these processes, improving tissue and organ survival and function. These have involved the use of preservation solutions. Although most research involves organ transplantation, these principles have been translated and applied to various tissues, surgical flaps and microvascular replantations. These studies have generally used the skin flap as their model with little knowledge regarding muscle flaps, the most vulnerable to the ischaemic process. This study targets the use of one such preservation system and uses skeletal muscle as its tissue model. The vascular anatomy of the rectus femoris muscle in the New Zealand white rabbit was studied anatomically and radiologically and thus described. The isolated rectus femoris muscle flap was harvested and perfused in-vitro with cooled, oxygenated University of Wisconsin solution (UWS) using a pulsatile renal perfusion pump. UWS was selected as it contains vital additives important in cryopreservation of organs. Monitoring of various physiological parameters was performed. The muscle was examined at 0, 4, 8, 12, 18 and 24 hours of extra-corporeal perfusion using warm and cold, non-perfused controls. The contralateral muscle served as the control. End-points were the percentage of muscle survival, as determined by a new grading system of muscle ischaemia, based on 3 light and 7 electron microscopic criteria. The overall percentage of muscle survival (combined light and electron microscopy scores) resulted in approximately 58% survival at 24 hours for the perfused muscle versus 31% for the cold stored muscle. The stored muscle had the same survival rate at 12 hours as did the perfused muscle at 24 hours. For all time periods beyond 4 to 8 hours, perfused muscle showed statistically improved survival rates compared to the stored muscle. Eight hours appears to be a crucial point beyond which survival in muscle deteriorates to a much greater degree without perfusion. Questions remain as to which method of preservation yields the best survival benefit and, as yet, there is no “ideal” perfusate. The future involves manipulating perfusion solutions and trying to arrest or reverse established warm ischaemia. Success of free tissue transfers and replantations of musclecontaining body parts may be enhanced. These techniques may also allow us to effectively store previously harvested flaps and eventually, to enter the realm of “banked” allograft tissue flaps.

The fundamental requirement of the research reported within this thesis is the provision of physical models to enable model based simulation of mainstream printed circuit assembly (PCA) process discrete events for use within to-be-developed (or under development) software tools which codify cause & effects knowledge for use in product and process design optimisation. To support a national competitive advantage in high reliability electronics UK based producers of aircraft electronic subsystems require advanced simulation tools which offer model based guidance. In turn, maximization of manufacturability and minimization of uncontrolled rework must therefore enhance inservice sustainability for 'power-by-the-hour' commercial aircraft operation business models.

My research broadly covers various important aspects of microfluidic devices and biosensors. Specifically, this dissertation reports: (1) a new and effective room temperature method of bonding polydimethylsiloxane (PDMS) microfluidics to substrates such as silicon and glass, (2) a new microfluidic pump concept and implementation specifically designed to repeatedly drive a small sample volume (<1 µL) very rapidly (~500 µL/min) through a sensor-containing flow channel to significantly decrease sensor response time through advection-driven rather than diffusion-driven mass transport, (3) use of a new microfluidic material based on polyethylene glycol diacrylate (PEGDA) to implement impedance-based dynamic nanochannel sensors for protein sensing, and (4) an investigation of galvanoluminescence and how to avoid it for conditions important to fluorescence-based dielectrophoresis (DEP) microfluidic biosensors. Over the last decade, the Nordin research group has developed a lab-on-a-chip (LOC) biosensor based on silicon photonic microcantilever arrays integrated with polydimethylsiloxane (PDMS) microfluidics for protein biomarker detection. Integration requires reliable bonding at room temperature with adequate bond strength between the PDMS element and microcantilever sensor substrate. The requirement for a room temperature process is particularly critical because microcantilevers must be individually functionalized with antibody-based receptor molecules prior to bonding and cannot withstand significant heating after functionalization. I developed a new room temperature bonding method using PDMS curing agent as an intermediate adhesive layer. Two curing agents (Sylgard 184 and 182) were compared, as well as an alternate UV curable adhesive (NOA 75). The bond strength of Sylgard 184 was found to be stronger than Sylgard 182 under the same curing conditions. Overnight room temperature curing with Sylgard 184 yields an average burst pressure of 433 kPa, which is more than adequate for many PDMS sensor devices. In contrast, UV curable epoxy required a 12 hour bake at 50 °C to achieve maximum bond strength, which resulted in a burst pressure of only 124 kPa. In many biosensing scenarios it is desirable to use a small sample volume (<1 µL) to detect small analyte concentrations in as short a time as possible. I report a new microfluidic pump to address this need, which we call a reflow pump. It is designed to rapidly pump a small sample volume back and forth in a flow channel. Ultimately, the flow channel would contain functionalized sensor surfaces. The rapid flow permits use of advection-driven mass transport to the sensor surfaces to dramatically reduce sensor response times compared to diffusion-based mass transport. Normally such rapid flow would have the effect of decreasing the fraction of analyte molecules in the volume that would see the sensor surfaces. By configuring the pump to reflow fluid back and forth in the flow channel, the analyte molecules in the small sample volume are used efficiently in that they have many opportunities to make it to the sensor surfaces. I describe a 3-layer PDMS reflow pump that pumps 300 nL of fluid at 500 µL/min for 15 psi actuation pressure, and demonstrate a new two-layer configuration that significantly simplifies pump fabrication. Impedance-based nanochannel sensors operate on the basis of capturing target molecules in nanochannels such that impedance through the nanochannels is increased. While simple in concept, the response time can be quite long (8~12 hours) because the achievable flow rate through a nanochannel is very limited. An approach to dramatically increase the flow rate is to form nanochannels only during impedance measurements, and otherwise have an array of nanotrenches on the surface of a conventional microfluidic flow channel where they are exposed to normal microfluidic flow rates. I have implemented such a dynamic nanochannel approach with a recently-developed microfluidic material based polyethylene glycol diacrylate (PEGDA). I present the design, fabrication, and testing of PEGDA dynamic nanochannel array sensors, and demonstrate an 11.2 % increase in nanochannel impedance when exposed to 7.2 µM bovine serum albumin (BSA) in phosphate buffered saline (PBS). Recently, LOC biosensors for cancer cell detection have been demonstrated based on a combination of dielectrophoresis (DEP) and fluorescence detection. For fluorescence detection it is critical to minimize other sources of light in the system. However, reported devices use a non-noble metal electrode, indium tin oxide (ITO), to take advantage of its optical transparency. Unfortunately, use of non-noble metal electrodes can result in galvanoluminescence (GL) in which the AC voltage applied to the electrodes to achieve DEP causes light emission, which can potentially confound the fluorescence measurement. I designed and fabricated two types of devices to examine and identify conditions that lead to GL. Based on my observations, I have developed a method to avoid GL that involves measuring the impedance spectrum of a DEP device and choosing an operating frequency in the resistive portion of the spectrum. I also measure the emission spectrum of twelve salt solutions, all of which exhibited broadband GL. Finally, I show that in addition to Au, Cr and Ni do not exhibit GL, are therefore potentially attractive as low cost DEP electrode materials.

Pour ce travail de thèse, nous avons réalisé cinq études originales en utilisant trois technologies d'imageries cardiovasculaires non-invasives.-Nous avons démontré, à partir d'une étude ex-vivo sur des artères coronaires humaines, que le scanner64 détecteurs ne permettait pas de caractériser précisément les différents composants des plaques. Ladistinction des plaques fibreuses et des plaques lipidiques est en effet impossible. Par ailleurs, notretravail a montré que l'IVUS ne devait pas servir d'imagerie de référence lors des études sur la plaque carcet examen présente lui aussi de nombreuses imprécisions.-Notre travail sur la thrombo-aspiration rapporte un effet très significatif de l'extraction effective dethrombus lors des thrombo-aspirations à la phase aiguë des STEMI, avec une réduction de la taille du no-reflow et de l'infarctus, évaluées en IRM ; une thrombo-aspiration positive représentant par ailleurs dansnotre travail, un critère indépendamment lié à la taille finale de l'infarctus. L'extraction effective dethrombus pourrait être considéré, en salle de cathétérisme, comme un critère de jugement de l'efficacitéde la thrombo-aspiration.-Notre étude sur les syndomes coronaires à coronaires angiographiquement normales a confirmé l'intérêtde l'IRM dans le bilan étiologique de cette présentation clinique, permettant un diagnostic étiologiquedans 2/3 des cas. Par ailleurs, nous avons observé une excellente évolution pour le tiers des patients chezqui l'IRM ne décèle pas d'anomalie myocardique. Des études d'une plus grande envergure serontnécessaires afin de confirmer nos résultats.-A partir d'IRM cardiaque réalisées chez des patients ayant présenté un premier épisode de STEMI, nousavons pu déterminer une valuer seuil de troponine prédictive de la survenue d'un no-reflow.-Enfin, à partir d'analyses en Speckle Tracking, nous avons mis en évidence une dysfonction systolique VD,objectivée par une altération des valeurs de Strain longitudinal VD, chez les patients présentant une EPgrave ou de gravité intermédiaire, comparativement à un groupe de patients avec une EP non grave. / In this doctoral thesis, we report on five original studies that use three différent non-invasive cardiovascular imaging techniques:- In an ex vivo study of human coronary arteries, we show that 64-slice computed tomography (CT) scan isnot capable of distinguishing between différent components of plaques. Indeed, it is impossible todifferentiate between fibrous and lipid plaques. Our study also showed that intravascular ultrasound(IVUS) should not be used as thé référence method in studies of plaque composition, since this techniquealso suffers from numerous limitations.- Our study of thé efficacy of thrombo-aspiration showed a significant benefit with effective extraction ofthrombus during thrombo-aspiration at thé acute phase of ST élévation myocardial infarction (STEMI),notably with a réduction of thé extent of no-reflow and of infarct size as evaluated by magnetic résonanceimaging (MRI). Productive thrombo-aspiration was shown in our study to be an independent predictor offinal infarct size. Effective extraction of thrombotic material could be considered in thé cathlab as acriterion for evaluating thé success of thé thrombo-aspiration procédure.- Our study of acute coronary syndromes with normal coronary arteries confirmed thé utility of MRI inestablishing thé etiology of this clinical présentation, and made it possible to establish an etiologicaldiagnosis in two-thirds of patients. We also observed excellent outcomes in thé third of patients in whomMRI did not find any myocardial anomalies. Larger studies are warranted to confirm thèse findings.- Based on cardiac MRI performed in patients presenting a first épisode of STEMI, we established athreshold value of troponin that predicts thé occurrence of no-reflow.- Lastly, using speckle-tracking analysis, we demonstrated impaired systolic right ventricular function inpatients with intermediate to high risk pulmonary embolism (PE), evaluated by altérations in longitudinalstrain values at thé level of thé right ventricle, compared to a control group of patients with low risk PE.

The factors determining the quality of surface mount technology (SMT) solder joints are numerous, and complex. The exploration of these factors, and how they may affect the reliability and quality of the joints can only be achieved through continuous research. In this project, essential areas of SMT joints were selected for study and analysis, with the intention of providing additional design and process guidelines for the production of quality SMT joints. In the infrared reflow process, one of the common defect phenomena is the occurrence of tombstoning; that is after soldering only one end of the component is soldered while the other is lifted up, assuming a position like a tombstone. The initiation of tombstoning during reflow was analysed based on the forces acting on the component. A model was developed to predict the initiation of this phenomenon. The model shows that, under vibration-free conditions, the surface tension of the molten solder is the source of the force causing the initiation of tombstoning. The contact angle, which varies with the length of the printed circuit board solder land, has a significant effect on the value of the surface tension acting as a force pulling upward on the component. The model further shows that tombstoning initiation is due to the combined effects of the surface tension; the weight of the component; the dimensions of the component; the length of the solder underneath the component; and the length of the solder protruding from the end of the component. Selected components were used as examples for predicting the conditions of initiation, and these conditions were further substantiated by a series of experiments. Another area of study was a method which directly pulled the components off printed circuit boards and this was used as a means for testing the bond quality of surface mount technology leadless chip solder joints. Components D7243, CC1206, RC1206, RC121O, and CC1 812 were selected for this study. It was found that the ultimate tensile force which breaks a component off the printed circuit board has the potential to be used as a parameter for measuring the quality of the solder joint. The effect of solder thickness on the strength of a joint has also been investigated. The shape of joints soldered by two methods, wave soldering and infrared reflow, were compared. Joints at the two ends of a component produced by infrared reflow were found more uniform than the ones produced by wave soldering. A recommendation is made here for the wave soldering approach in achieving uniform solder joints. The effects of solder shape on the joint strength were further investigated by finite element analysis. A convex joint was found marginally more robust than a concave joint. Two aspects of the internal structure of SMT solder joints were investigated, void content and copper/tin intermetallic compounds. The voiding conditions of wave-soldered and infrared reflow joints were compared. No voids were found in all specimens that were produced by wave soldering. However, there were always voids inside joints produced by infrared reflow. Microhardness tests indicated that the hardness of compounds at the copper/solder interface of infrared reflowed joints is lower than that in the wave-soldered joints. It is considered that the lower hardness of the interfacial region of the infrared reflowed joints is due to the presence of voids. Scanning electron microscopy was used to study the formation of copper/tin intermetallic compounds for joints produced by infrared reflow. The results show that Cu 6 Sn5 was the only compound with a detectable thickness. Other compounds such as Cu3 Sn, were virtually not found at all. Aging of the joints at 100°C, shows that both the Cu 6Sn5 and the overall interfacial thickness grew with time. One of the important areas which had been overlooked previously and was studied in some details was the effects of solder paste exposure on the quality of solder paste. The characteristic changes of solder paste due to exposure were investigated in three areas, weight loss, tackiness, and rheology. The evaporation of low boiling point solvents was considered as the main contribution to the loss in the weight of the solder paste. The weight loss against exposure time was found to follow an exponential behaviour. A method was designed to evaluate the tackiness changes of solder paste due to exposure. It was found that the decay of tackiness against exposure time can be expressed by a power law. It is recommended that solder paste manufacturers should provide the necessary characteristic constants so as to enable the characteristics to be calculated after a specific exposure. The rheological changes of the solder paste as a result of exposure were also investigated. The implication on the printability of the solder paste due to these changes was studied and discussed.

Pour ce travail de thèse, nous avons réalisé cinq études originales en utilisant trois technologies d'imageries cardiovasculaires non-invasives.-Nous avons démontré, à partir d'une étude ex-vivo sur des artères coronaires humaines, que le scanner64 détecteurs ne permettait pas de caractériser précisément les différents composants des plaques. Ladistinction des plaques fibreuses et des plaques lipidiques est en effet impossible. Par ailleurs, notretravail a montré que l'IVUS ne devait pas servir d'imagerie de référence lors des études sur la plaque carcet examen présente lui aussi de nombreuses imprécisions.-Notre travail sur la thrombo-aspiration rapporte un effet très significatif de l'extraction effective dethrombus lors des thrombo-aspirations à la phase aiguë des STEMI, avec une réduction de la taille du no-reflow et de l'infarctus, évaluées en IRM ; une thrombo-aspiration positive représentant par ailleurs dansnotre travail, un critère indépendamment lié à la taille finale de l'infarctus. L'extraction effective dethrombus pourrait être considéré, en salle de cathétérisme, comme un critère de jugement de l'efficacitéde la thrombo-aspiration.-Notre étude sur les syndomes coronaires à coronaires angiographiquement normales a confirmé l'intérêtde l'IRM dans le bilan étiologique de cette présentation clinique, permettant un diagnostic étiologiquedans 2/3 des cas. Par ailleurs, nous avons observé une excellente évolution pour le tiers des patients chezqui l'IRM ne décèle pas d'anomalie myocardique. Des études d'une plus grande envergure serontnécessaires afin de confirmer nos résultats.-A partir d'IRM cardiaque réalisées chez des patients ayant présenté un premier épisode de STEMI, nousavons pu déterminer une valuer seuil de troponine prédictive de la survenue d'un no-reflow.-Enfin, à partir d'analyses en Speckle Tracking, nous avons mis en évidence une dysfonction systolique VD,objectivée par une altération des valeurs de Strain longitudinal VD, chez les patients présentant une EPgrave ou de gravité intermédiaire, comparativement à un groupe de patients avec une EP non grave. / In this doctoral thesis, we report on five original studies that use three différent non-invasive cardiovascular imaging techniques:- In an ex vivo study of human coronary arteries, we show that 64-slice computed tomography (CT) scan isnot capable of distinguishing between différent components of plaques. Indeed, it is impossible todifferentiate between fibrous and lipid plaques. Our study also showed that intravascular ultrasound(IVUS) should not be used as thé référence method in studies of plaque composition, since this techniquealso suffers from numerous limitations.- Our study of thé efficacy of thrombo-aspiration showed a significant benefit with effective extraction ofthrombus during thrombo-aspiration at thé acute phase of ST élévation myocardial infarction (STEMI),notably with a réduction of thé extent of no-reflow and of infarct size as evaluated by magnetic résonanceimaging (MRI). Productive thrombo-aspiration was shown in our study to be an independent predictor offinal infarct size. Effective extraction of thrombotic material could be considered in thé cathlab as acriterion for evaluating thé success of thé thrombo-aspiration procédure.- Our study of acute coronary syndromes with normal coronary arteries confirmed thé utility of MRI inestablishing thé etiology of this clinical présentation, and made it possible to establish an etiologicaldiagnosis in two-thirds of patients. We also observed excellent outcomes in thé third of patients in whomMRI did not find any myocardial anomalies. Larger studies are warranted to confirm thèse findings.- Based on cardiac MRI performed in patients presenting a first épisode of STEMI, we established athreshold value of troponin that predicts thé occurrence of no-reflow.- Lastly, using speckle-tracking analysis, we demonstrated impaired systolic right ventricular function inpatients with intermediate to high risk pulmonary embolism (PE), evaluated by altérations in longitudinalstrain values at thé level of thé right ventricle, compared to a control group of patients with low risk PE.

This thesis deals with the teoretical description of the solder during reflow soldering the solder paste and describing the defects arising during this process. Practical part of the thesis describes the testing of two solder pastes with different silver content before and after the accelerated temperature cycle.