The fast growth of the Chinese oldest old population indicates higher demand for long-term care. In China, families assume the primary responsibility of caring for older adults. Since the oldest elders are more likely to be widowed, their adult children usually become their caregivers. Focusing on the Chinese adult children who provide care for their oldest-old parents, this study documented and helped to explain Chinese adult children's caregiving strain. A conceptual framework was developed based on Pearlin's stress process theory, Higgins' framework of self-concept discrepancy, and previous studies on family caregivers of elders. Using an existing dataset from the 2005 Chinese Longitudinal Healthy Longevity Survey of 895 caregivers and their care recipients, the researcher tested whether and how caregiving context (caregiver's structured context and care recipients' needs for care), caregiving performance, and sibling support were related to five types of caregiving strain including sacrifice strain, exhaustion strain, capability strain, expectation strain, and dependency strain. The results indicate that caregiving context and caregiving performance are statistically related to different types of caregiver strain. Three independent variables in the set of caregiving context, self-evaluation of living standard, education, and cultural identity, were related to two types of caregiver strain in different directions. The caregivers who were the eldest sons, who were females caring for female elders, who had a close relationship with their care recipients, who lived with the care recipients, who provided care for the elders with more needs for care in ADL (Activities of Daily Living), or whose care recipients had health insurance reported higher levels of at least one type of caregiver strain. Care recipients' cognitive status and entitlement to pension were negatively related to at least one type of caregiver strain. Caregivers' rural residence, having a job outside the family, having a child under age 16, and care recipient's needs for care in IADL (Instrumental Activities of Daily Living) were not found to be related to any type of caregiver strain. Monetary assistance, which was indicated by the proportion of their annual household per capita income that the caregivers provided to care recipients, was found to be positively related to caregivers' capability strain. The amount of time spent in caregiving (time assistance) was positively related to three types of caregiver strain: exhaustion, expectation, and dependency strain. Time assistance was also found to mediate the relationship between care recipients' needs for care in ADL and caregivers' exhaustion strain and the relationship between dependency strain and three caregiving context variables: closeness between caregivers and care recipients, co-residence with care recipients, and care recipients' needs for care in ADL. The results revealed the importance of caregiving context and caregiving performance in explaining Chinese adult-child caregivers' experience and the necessity of investigating caregiver strain in different dimensions. This study contributes to understanding caregiver strain from a filial perspective. The results imply directions for future research, social work practice and education, and policy legislation in addressing Chinese adult children's strain in caring for their oldest-old parents.

This research was done as part of a long term project, with the goal to monitor multiple bridges over an extended period of time. Due to the nation’s aging infrastructure and the limited amount of funds to upgrade and maintain it, structural health monitoring (SHM) is very important because it provides in depth information about a structure to be used in decision making. SHM of bridges includes monitoring the effects of traffic loads. This paper discusses the development of a bridge weigh-in-motion (B-WIM) technique that uses the rainflow counting of strain cycles. Typical B-WIM techniques have proven to be accurate but require large algorithms and gauges at multiple locations across the span, and the strain gauge temperature drift must be accounted for. The rainflow B-WIM (RF-BWIM) decreases the processing of the B-WIM and automatically accounts for drift, thus allowing temperature and other analyses of the same bridge to be possible. RF-BWIM also has the potential to decrease the number of sensors required. Strain data taken from an existing long term monitoring system was used to develop the RF-BWIM. The development of the RF-BWIM, as well as a method to determine a virtual gross vehicle weight (C-GVW) used in calculating the RF-BWIM output, is presented.

In 1993, to determine if wild-strain mallard releases could be used as a management practice to increase local mallard breeding populations, I released 2,344,4.5- week-old mallard ducklings (1,200 females and 1,144 males) to wetlands on 12,10.4-km2 sites (approximately 200 per site, 100 females, 100 males) in the North Dakota Prairie Pothole Region. I monitored the release sites to determine if any relationship existed between site characteristics and time of release to duckling survival estimates. I conducted breeding pair surveys during 1994 and 1995 on treatment and paired control sites to compare post-release population levels. Lastly, I analyzed return data and habitat use, and conducted behavioral experiments to determine if wild-strain mallards experienced higher mortality rates and if any observed differences could be explained by behavior. In 1994,1 observed 55 of the nasal saddled ducklings returning as adult fem ale to the release sites. In 1995, only 5 nasal saddled females were observed, both on treatment and control sites. No difference was observed hi breeding pair populations on treatment and control sites in 1994 (P = 0.18) and 1995 (P = 0.59). Hard-released wild-strain mallard females had lower survival rates than wild (P = 0.01) and modified gentle-release wild-strain females (P = 0.05). Ail wild-strain females were virtually eliminated from the population by year 4. This suggests that these buds may have been more vulnerable to predation and other mortality factors than wild females. Breeding wild and wild-strain mallard females reacted similarly to human approach, but when flushed, wild females flew farther than wild-strain females (P = 0.0002). Wetlands used by wild-strain females differed from wild females during breeding by type (P < 0.0001) and cover (P = 0.0003) classification. Wild-strain females selected larger, more permanent wetlands exhibiting less emergent vegetation than did wild counterparts. These differences may help to explain why wiki-strain mallard releases did not increase local breeding populations. The lack of band recoveries for wild-strain females during the latter years when viewed in the context of the observed behavioral differences suggests that these birds were unable to adapt to conditions in the wild.

Current two dimensional face recognition methods rely on visible photometric or geometric attributes that are present in the intensity image. In many of these approaches a technique called Principal Component Analysis (PCA) is extensively used. PCA extracts the maximum intensity variations from the set of input images in the form of "eigen" faces which are used as a feature vector. In these approaches the intensity images used were mostly that of the subject's frontal face, which yielded promising results after doing PCA. These approaches however fail in the presence of facial expression, unstable lighting conditions and artifacts such as make-up, glasses etc. Thus, it is desirable to establish a new biometric source that will be least affected bythe afore mentioned factors. This study describes a face recognition method that is designed based on the consideration of anatomical and biomechanical characteristics of facial tissues. During facial expressions such as smile, frown, anger etc, various muscles get activated in tandem. A strain pattern inferred from a face expression can reveal an individual's signature associated with the underlying anatomical structure, and thus has the potential for face recognition. In this study, the strain is computed by measuring the displacement of a point on the face that results from a facial expression such as opening the mouth. The information provided by the change in the depth value for the face across the open and close mouth frames does not provide any information required for computing the strain maps, because the strain map depends on the relative displacements of two points on the face, which remains same with rigid motions of the face such as rotation and translation. Hence the information in the 2D spaceis sufficient to compute strain since the depth is assumed constant. The approach used to calculate strain computes the strain distribution directly using the mathematical definition of strain as the derivative of displacement in 2D space (XY plane). The strain values obtained are converted to gray scale intensity images, which are used as inputs for the intensity based PCA analysis. Experiments were conducted using 62 subjects. The data set comprised of two pairs of images for a subject: closed mouth and open mouth under bright and low light. Analysis of CMC and ROC curves indicate that the proposed strain map biometric is a promising new biometric that has the potential to improve the performance of current face recognition method. In summary, the contribution of this thesis is twofold: 1. Facial strain map proves to be promising new biometric. 2. Strain map helps increase the identification rate when used in conjunction with intensity based biometric as a multi-classifier.

General strain theory has received a fair amount of empirical support and theoretical elaboration over the past several years. Since the introduction of general strain theory, Agnew and others have attempted to increase the comprehensiveness of the processes involved in strain theory. Until recently, the general strain theory literature has ignored what Agnew and associates (Agnew, Brezina, Wright, & Cullen, 2002) argue may be one of the most important conditioning effects of the strain-crime relationship, namely the dispositions or personality traits of the individual experiencing strain. Recently, Agnew and associates (2002) published results from a study examining the conditioning effects of personality traits (i.e., negative emotionality and low constraint) on the strain-delinquency relationship. Their findings indicated that certain personality traits significantly condition the effect of strain on delinquency. Research has suggested that more severe personality and behavioral traits, such as psychopathy, also influence criminality. The present study examined moderating effects of both personality dispositions and psychopathic behavioral features among a sample of 137 youths referred to juvenile diversion by the court system. The results suggest that personality dispositions and psychopathic behavioral features do not significantly moderate the strain-delinquency relationship. In addition, this study conducted ad hoc analyses examining whether or not delinquency significantly increases the likelihood that subsequent strain and delinquency will result (i.e., a state dependence explanation (see Nagin & Farrington, 1992; Nagin & Paternoster, 1991)). Moderating effects of personality and psychopathy were also included in this model. Further, the role of strain as a mediator for the personality and psychopathy link to delinquency was tested. The findings suggest that delinquency exacerbated subsequent strain and delinquency levels among these youths. Personality and psychopathic features did not moderate the strain-delinquency relationship. Strain did not significantly moderate the personality-delinquency relationship. Limitations and implications for future research and policy are discussed.

Les travaux de cette thèse ont porté sur le développement d’une nouvelle loi de comportement hyperélastique, isotrope et incompressible permettant de modéliser les matériaux caoutchouteux en grande déformation et en grand déplacement. Cette nouvelle loi combine une approche moléculaire et une approche phénoménologique, ce qui permet de couvrir un spectre large de sollicitations. Elle est constituée par la superposition de quatre termes :– un terme lié à la contrainte d’entrelacement des chaînes macromoléculaires observée avec le phénomène de cristallisation. Ce terme est modélisé par une fonction logarithmique provenant de l’énergie phénoménologique de Gent-Thomas,– un terme lié à l’hypothèse des déformations affines observées avec le raidissement final de certaines chaînes macromoléculaires des élastomères. Ce terme provient de la probabilité non-Gaussienne de Langevin. Nous l‘avons modélisé par la loi moléculaire 8-chaines d’Arruda-Boyce avec un aménagement qui consiste à utiliser une approximation originale de la fonction de Langevin inverse,– un terme lié à la contrainte des chaînes ayant des déformations non-affines. Ce terme est modélisé par une fonction Gaussienne sous forme intégrale. Il s’agit de l’une des contributions originale de ce travail de thèse,– une partie volumique standard permettant de prendre en compte l’incompressibilité du matériau.Les deux principales originalités de la thèse concernent donc l’élaboration d’une approximation inédite de la fonction de Langevin inverse ainsi que la construction d’une nouvelle densité d’énergie hyperélastique isotrope, incompressible et hybride.Afin d’étudier la pertinence du modèle proposé, des comparaisons ont été réalisées avec plusieurs jeux de données expérimentales disponibles dans la littérature. Ces comparaisons ayant été couronnées de succès, l’implémentation numérique du modèle que nous proposons a été effectuée dans le code universitaire aux éléments finis FER. / This thesis concerns the development of a new incompressible isotropic hyperelastic behavior law allowing the modeling of rubber materials with large strain and large displacement. This new law mixes a molecular approach with a phenomenological one and therefore covers a wide range of loading. It has been built by a sum over four terms:– a term related to the interleaving macromolecular chains observed with the crystallization phenomenon. This term is modeled by a logarithmic function coming from the phenomenological energy of Gent-Thomas,– a term related to the assumption of affine deformations observed with the final stiffening of a part of macromolecular elastomeric chains. This term comes from the non Gaussian probability of Langevin. We have modeled it by the 8-chains molecular law of Arruda-Boyce but with an original approximation of the inverse of the Langevin function,– a term related to the stress occurring with non affine strains. This term has been modeled by a Gaussian function adopting an integral form. This is one of the original contribution of this thesis work,– a classical volumetric term taking into account the incompressibility of the material.The two main originalities of the thesis are therefore the introduction of a new approximation of the inverse of the Langevin function and the development of a new hyperelastic energy density which is isotropic, incompressible and hybrid.In order to study the efficiency of the proposed model, comparisons were made with several experimental data available in the literature. These comparisons have been successful and we have implemented our model in the university finite element software FER.

While support is generally a helpful resource for employees, support can also serve as a job stressor. Unhelpful workplace social support (UWSS) is any action taken by a supervisor and/or colleague that is intended to benefit another worker but is perceived as unhelpful or harmful by the recipient. A series of three studies identified types of UWSS, developed a measure of UWSS, and established a nomological network of variables related to UWSS. In Study 1, critical incidents were collected from 116 employees, and a content analysis revealed 11 distinct categories of UWSS. A measure of UWSS was developed in Study 2, and a nomological network of variables related to the construct was examined using responses from 176 employees. Results demonstrate that UWSS is associated with higher negative affect, lower competence-based self-esteem, lower coworker satisfaction, higher work-related burnout, higher organizational frustration, and higher physical symptoms (e.g., headache, nausea, and fatigue) among recipients. Study 3 replicated the findings using data from 496 registered nurses to mitigate the chances of reporting Type 1 errors. Together, the studies demonstrate that unhelpful workplace social support is a meaningful job stressor worthy of further investigation.

Workers are exposed to stressful thermal work environments in multiple industries every day. Methods for assessing heat stress often struggle to balance productivity without compromising the health of the workers. The Thermal Work Limit (TWL) is a method that has been adopted in areas outside of the United States as a viable method for heat assessment that combines health with productivity. TWL recommends a maximum metabolic rate for a given set of environmental conditions, clothing ensemble and acclimatization state. The purpose of this paper was to evaluate the validity of the TWL against a set of heat stress data known to be at the maximum sustainable level. A range of conditions were combined through environmental (20%, 50% and 70% relative humidity), clothing (woven clothing, WC; particle barrier coveralls, PB; water barrier coveralls, WB; and vapor barrier coveralls, VB), and workload factors (metabolic rates at low, L; moderate, M; and high, H) at the transition from sustainable to unsustainable exposure to ensure that the TWL method is thoroughly explored. Data from previous heat stress studies were used to compare the difference in predicted TWL with a calculated value. An analysis of variance (ANOVA) demonstrated that there were significant effects of the TWL due to clothing, metabolic rate level and relative humidity level. TWL provided similar results for WC, PB and WB, but had systematically lower values for VB. This suggested a more protective recommendation with high evaporative resistance. As the metabolic rate increased, the recommended limiting TWL also went up out of proportion to the metabolic rate, which provided greater protection at increasing metabolic rates. Under drier conditions (20% relative humidity), the TWL was systematically lower than for 50% and 70% relative humidity. While there were significant differences due to the main effects, the TWL was designed to be used without defined limits on environmental conditions, metabolic rate or clothing. Therefore, all of the conditions represented a comprehensive test of the TWL. Overall, the TWL was less protective than the current methods used by ACGIH Threshold Limit Values (TLV) and National Institute for Occupational Safety and Health (NIOSH) Recommended Exposure Limit (REL). At the threshold, the TWL had a 7% probability of being unsustainable compared to the threshold probability of 1% for the TLV and REL.

The Staphylococcus aureus two component system (TCS) sae governs expression of numerous virulence factors, including Eap (extracellular adherence protein), which in turn among other functions also mediates invasion of host cells. The sae TCS is encoded by the saePQRS operon, with saeS coding for the sensor histidine kinase (SaeS) and saeR encoding the response regulator (SaeR). The saeRS system is preceded by two additional open reading frames (ORFs), saeP and saeQ, which are predicted to encode a lipoprotein (SaeP) and a membrane protein (SaeQ), respectively. Earlier, we have shown that SDS-containing subinhibitory concentrations of biocides (Perform®) and SDS alone activate sae transcription and increase cellular invasiveness in S. aureus strain Newman. The effect is associated with an amino acid exchange in the N-terminus of SaeS (L18P), specific to strain Newman. In this work, the role of whether the two additional genes, saePQ coding for the accessory proteins SaeP and SaeQ, respectively, are involved in SDS-mediated saeRS was investigated. It could demonstrated that the lack of the SaeP protein resulted in an increased saeRS transcription without SDS stress in both SaeSL/P variants, while the SDS effect was less pronounced on sae and eap expression compared to the Newman wildtype, suggesting that the SaeP protein represses the sae system. Also, SDS-mediated inductions of sae and eap transcription along with enhanced invasion were found to be dependent on presence of the SaeSP variant in Newman wildtype. On the other hand, the study also shows that the saePQ region of the sae operon is required for fully functional two-component system saeRS under normal growth conditions, but it is not involved in SDS-mediated activation of the saeS signaling and sae-target class I gene, eap. In the second approach, the study investigates whether SDS-induced sae expression and host cell invasion is common among S. aureus strains not carrying the (L18P) point mutation. To demonstrate this strain Newman, its isogenic saeS mutants, and various S. aureus isolates were analysed for sae, eap expression and cellular invasiveness. Among the strains tested, SDS exposure resulted only in an increase of sae transcription, Eap production and cellular invasiveness in strain Newman wild type and MRSA strain ST239-635/93R, the latter without an increase in Eap. Interestingly, the epidemic community-associated MRSA strain, USA300 LAC showed a biphasic response in sae transcription at different growth stages, which, however, was not accompanied by increased invasiveness. All other clinical isolates investigated displayed a decrease of the parameters tested. While in strain Newman the SDS effect was due to the saeSP allele, this was not the case in strain ST239-635/93R and the biphasic USA300 strains. Also, increased invasiveness of ST239-635/93R was found to be independent of Eap production. Furthermore, to investigate the global effect of SDS on sae target gene expression, strain Newman wild-type and Newman ∆sae were treated with SDS and analyzed for their transcription profiles of sae target genes using microarray assays. We could show that subinhibitory concentrations of SDS upregulate and downregulate gene expression of several signaling pathways involved in biosynthetic, metabolic pathways as well as virulence, host cell adherence, stress reponse and many hypothetical proteins. In summary, the study sheds light on the role of the upstream region saePQ in SDS-mediated saeRS and eap expression during S. aureus SDS stress. Most importantly, the study also shows that subinhibitory SDS concentrations have pronounced strain-dependent effects on sae transcription and subsequent host cell invasion in S. aureus, with the latter likely to be mediated in some strains by other factors than the known invasin Eap and FnBP proteins. Moreover, there seems to exist more than the saeSP-mediated mechanism for SDS-induced sae transcription in clinical S. aureus isolates. These results help to further understand and clarify virulence and pathogenesis mechanisms and their regulation in S. aureus. / Das Zwei Komponenten-Systems (TCS) Sae in S. aureus reguliert die Expression einer Vielzahl von Virulenzfaktoren, dazu gehört unter anderem das extrazelluläre Adhärenzprotein Eap, welches neben weiteren Funktionen, die Invasion in eukaryotische Wirtszellen vermittelt. Die Gene des sae TCS sind in einem Operon organisiert (saePQRS), wobei saeS für die sensorische Histidinkinase (SaeS) und saeR für den „Response Regulator“ (SaeR) kodieren. Diesen Genen sind zwei weitere Genabschnitte, saeP und saeQ, vorangestellt, wobei saeP vermutlich für ein Lipoprotein (SaeP) und saeQ für ein Membranprotein (RelQ) kodieren. In einer früheren Arbeit konnten wir zeigen, dass SDS-haltige Biozide (Perform©) unter sub- inhibitorischen Konzentrationen, sowie reines SDS, die sae Transkription aktiviert und dadurch zu einer erhöhten Invasion des S. aureus Stamms Newman in Wirtszellen führt. Dieser Effekt ist assoziiert mit einem spezifischen Aminosäureaustausch im N-terminus von SaeS (L18P) des Stamm Newman. In dieser Arbeit soll nun die Beteiligung der zwei zusätzlichen Gene, saeP und saeQ, an der SDS vermittelten transkriptionellen Induktion von saeR/S untersucht werden. Es konnte gezeigt werden, dass ohne SaeP, die saeR/S Transkription in beiden SaeL/P Varianten erhöht war, wobei eine zusätzliche SDS Behandlung hierfür nicht notwendig war. Im Gegenteil, es zeigte sich, dass der SDS Effekt auf die sae und eap Expression in der saeP Mutante deutlich weniger ausgeprägt ist als im Wildtyp Stamm. Das läßt vermuten, dass das Lipoprotein SaeP repremierend auf das sae System einwirkt. Des Weiteren wurde festgestellt, dass die SDS vermittelte transkriptionelle Induktion von sae und eap, zusammen mit der erhöhten Invasion, abhängig vom vorhanden sein der SaeSP Variante im Newman Wildtyp Stamm ist. Die Arbeit zeigt, dass die saePQ Region wichtig ist für die vollständige Funktion des Zwei Komponenten Systems SaeRS unter normalen Wachstumsbedingungen. Jedoch ist diese Region nicht involviert in der Aktivierung von SaeS, mit SDS als Signalgeber, sowie der darauffolgenden Aktivierung des sae Zielgens eap. In einem zweiten Ansatz wurde untersucht, ob die SDS induzierte sae Expression und Wirtszellinvasion auch häufig in S. aureus Stämmen auftritt, welche keine (L18P) Punktmutation besitzen. Dafür wurde Stamm Newman, die isogene saeS Mutante und verschiedene S. aureus Klinikisolate auf ihre sae, eap Expression, sowie zelluläre Invasionsfähigkeit hin analysiert. Von den getesteten Stämmen reagiert nur Wildtyp Stamm Newman und ein MRSA Stamm ST239-635/93R mit gesteigerter sae Transkription, Eap Produktion und zellulärer Invasion. Der MRSA Stamm jedoch ohne erhöhte Eap Produktion. Interessanterweise zeigt der „community- associated“ MRSA Stamm USA300 LAC eine biphasische sae Transkription in verschiedenen Wachstumsphasen, welche jedoch nicht einhergeht mit erhöhter Invasion. Alle anderen Klinikisolate zeigten abnehmende Tendenzen in den getesteten Parametern. Während im Stamm Newman der SDS Effekt auf das saeSP Allel zurückzuführen ist, gilt dies nicht für den Stamm ST239-635/93R, sowie den biphasischen Stamm USA300. Außerdem konnte gezeigt werden, dass die erhöhte Invasion des Stamms ST239-635/93R unabhängig von seiner Eap Produktion ist. Des Weiteren zeigten wir den globalen Effekt von SDS auf die sae Zielgenexpression. Dafür behandelten wir Wildtyp Stamm Newman mit SDS und analysierten die Transkription der sae Zielgene mittels Microarray Analyse. Wir konnten zeigen, dass subinhibitorische SDS Konzentrationen, induzierende als auch repremierende Auswirkungen auf die Genexpression haben. Dabei sind Gene betroffen, die involviert sind in verschiedene Signalwege, Biosynthese/Metabolismus als auch in Virulenz, Wirtzelladhärenz und Stressantwort. Zusammenfassend gibt die Arbeit Aufschluss über die Rolle der „upstream“ Region saePQ hinsichtlich der SDS-abhängigen saeRS und eap Expression in S. aureus. Am wichtigsten ist hierbei die Erkenntnis, das subinhibitorische SDS Konzentrationen einen deutlichen stammabhängigen Effekt auf die sae Transkription und daraus folgernd auf die Wirtszellinvasion von S. aureus haben. Letzteres wird vermutlich in manchen Stämmen durch andere Faktoren als die bekannten Invasinproteine Eap und FnBP vermittelt. Außerdem scheint es in den klinischen S. aureus Isolaten mehr als nur den saeSP abhängigen Mechanismus der sae Induktion durch SDS zu geben. Diese Ergebnisse helfen uns die Virulenz und pathogenen Mechanismen als auch deren Regulation in S. aureus zu verstehen. Die Beobachtungen tragen zu unserem Verständnis bei, wie das sae System Signale der Umgebung detektieren kann. Dies ist bis jetzt eine Fragestellung mit vielen Unbekannten.

Nous vous proposons un procédé de fabrication pour obtenir des dispositifs de graphène suspendu avec rendement élevé (~ 90%). Surtout, nous nous concentrons sur l'intégrabilité de procédé ainsi que sa compatibilité avec les technologies existantes de silicium. Pour répondre à ces questions, nous avons développé un système de fabrication à base de graphène qui était synthétisé par dépôt chimique en phase vapeur (CVD).L'étape la plus importante dans le processus de fabrication est liée à la gravure du substrat de SiO2 sous-jacente à suspendre les rubans de graphène. Il est souvent rapporté dans la littérature que, à ce stade, les forces capillaires peuvent provoquer l'effondrement des rubans de graphène. En dehors de cet effet, nous avons trouvé que la qualité de l'interface entre le masque de gravure et le substrat est essentielle pour suspendre les dispositifs de graphène avec succès. Ce n'est que lorsque la qualité de cette interface a été amélioré, nous avons atteint des rendements remarquablement élevés d'environ 90%. Caractérisation par spectroscopie Raman, la microscopie électronique à balayage (MEB) et microscopie à force atomique (AFM), qui était effectuée après chaque étape de fabrication, ont attesté que notre méthodologie n'a aucune dommage sur la qualité du graphène.Par la suite, nous avons utilisé la spectroscopie Raman pour étudier le dopage et la contrainte dans nos dispositifs de graphène CVD. Alors que nous avons observé une forte dopage de type p sur graphène supporté sur SiO2 dans l'air, le dopage seul ne peut pas tenir compte des spectres observés. Au lieu de cela, nous concluons que les échantillons de graphène mesurées présentent une contrainte interne de compression, qui ne se relâche pas complètement pendant la fabrication. Nous attribuons cette contrainte au budget de la température de CVD et au polymère rigide de transfert.Enfin, nous avons étudié les caractéristiques électriques de nos dispositifs à température ambiante ainsi qu'à basse température. Les mesures ont confirmé la forte dopage de type p de graphène, et en suite, ‘back-gating' ont donné une modulation faible de courant. Mesures magnéto-transport, qui sont effectuées à 20 K et 4, ont été utilisés pour extraire la densité de porteurs et la mobilité des dispositifs supportés sur SiO2. Faibles valeurs de mobilité sont attribuées à la diffusion par les joints de grains. À des champs magnétiques faibles, nous avons observé des signatures de localisation faible, ce qui implique que ‘intervalley scattering' est le mécanisme dominant dans nos échantillons. À des champs magnétiques élevés, la résistivité longitudinale a montré oscillations robuste à température qui pourraient être identifiés comme niveaux de Landau. / We propose a high-yield (~ 90%) fabrication route to obtain suspended graphene devices. Importantly, we focus on the scalability of the process as well as its compatibility with existing Si technologies. To address these issues, we developed a fabrication scheme based on graphene grown by chemical vapor deposition (CVD).The most crucial step in the fabrication process relates to the etching of the underlying SiO2 substrate to suspend the graphene ribbons. It is often reported in the literature that at this stage, capillary forces can lead to the collapse of the graphene beams. We have found that apart from this effect, the quality of the interface between the etch mask and the substrate is key to successfully suspend graphene devices. Only when the quality of this interface was improved, were we able to achieve remarkably high yields of approximately 90%. Characterization by Raman spectroscopy, scanning electron microscopy (SEM) and atomic force microscopy (AFM), performed after each step of fabrication, attested that our methodology does not impact the quality of the graphene.We have subsequently employed Raman spectroscopy to investigate doping and strain in our CVD graphene devices. While we did observe a strong p-type doping of graphene supported on SiO2 in air, doping alone cannot account for the observed spectra. Instead, we conclude that the measured graphene samples display a compressive internal strain, which does not fully relax during fabrication. We attribute this strain to the large temperature budget of the CVD process and to the rigid transfer polymer.