Les tapis de nanotubes de carbone verticalement alignés sont des candidats potentiels pour des applications telles que les interconnexions ou les matériaux d'interface thermique. Ce travail de recherche porte sur la synthèse de tapis de nanotubes de carbone alignés selon le procédé de dépôt chimique en phase vapeur (CVD) d'aérosols liquides, sur l'élaboration de nanocomposites constitués de différentes nuances de matrices époxy infiltrées au sein de ces tapis, ainsi que sur l'étude des propriétés mécaniques et thermiques longitudinales et transverses des tapis secs eux-mêmes et des nanocomposites 1D formés. Les conditions de synthèse permettent notamment de faire varier les caractéristiques des tapis telles que leur épaisseur, leur masse volumique, le diamètre externe moyen des nanotubes de carbone (NTC), l'espace intertube et la teneur volumique en NTC, alors que leur structure cristalline peut être modifiée par le biais d'un traitement thermique à haute température. L'objectif principal de ce travail consiste à démontrer et quantifier l'effet de certaines caractéristiques des tapis de nanotubes de carbone sur les propriétés mécaniques et thermiques des différents types de tapis et matériaux composites obtenus. Les deux méthodes d'imprégnation mises en oeuvre, voie liquide et infusion, conduisent à des tapis de NTC alignés denses avec un alignement des NTC conservé et une répartition homogène des NTC au sein du système époxy. La fraction volumique en NTC s'avère être le paramètre-clé permettant d'exacerber, dans la direction longitudinale aux NTC, les propriétés mécaniques et thermiques des nanocomposites. Par ailleurs, les tapis de NTC et les nanocomposites voient leurs propriétés de conduction thermique longitudinale nettement exacerbées lorsque les NTC présentent une amélioration de leur structure cristalline. L'augmentation significative des performances apportées par les tapis de NTC verticalement alignés au sein de ces matériaux nanocomposites anisotropes par rapport aux matrices organiques non chargées est prometteuse et ouvre des pistes de réflexion visant à répondre aux nouvelles exigences de multifonctionnalité des secteurs de l'aéronautique et de l'aérospatial. / Vertically aligned carbon nanotube carpets are potential candidates for applications such as interconnections or thermal interface materials (TIMs). This research work deals with the synthesis of aligned carbon nanotube carpets from the aerosol assisted chemical vapour deposition (CVD) technique, with the elaboration of nanocomposites made of different grades of epoxy matrix infiltrated within these carpets, as well as the study of both longitudinal and transverse mechanical and thermal properties of dry carpets themselves and 1D-nanocomposites separately. The synthesis conditions notably enable to vary characteristics of the differents carpets such as their thickness, their density, the mean external diameter of the carbon nanotubes (CNT), the intertube space and the CNT volume fraction, whereas their crystalline structure can be modified with a high temperature thermal treatment. The main goal of this work is to prove and quantify the effect of some of the characteristics of the carbon nanotubes carpets on both mechanical and thermal properties of the different kinds of CNT carpets and resulting composite materials. The two impregnation methods used, liquid way and infusion, lead to dense CNT carpets with a preserved alignment of the CNT and an homogeneous distribution of these latest within the epoxy system. The CNT volume content is evidenced as the key-parameter exacerbating the mechanical and thermal properties mainly in the longitudinal direction compared with the alignment axis of the CNTs. Moreover the mechanical and thermal conduction properties of the CNT carpets and the 1D-nanocomposites are clearly increased when the crystalline structure of the CNT is improved. The significant increasing of the properties brought by the vertically aligned CNT within these anisotropic 1D-nanocomposites compared with the only organic matrixes is promising and opens new pathways aiming to meet the latest specifications related to multifunctionnality in fields such as aeronautics and aerospace.

Abstract Carbon nanotubes (CNTs) have extraordinary mechanical and electrical properties. Together with their small dimensions and low density, they are attractive candidates for building blocks in future nanoelectromechanical systems and for many other applications. The extraordinary properties are however only attained by perfectly crystalline CNTs and quickly deteriorate when defects are introduced to the structure. The growth technique affects the crystallinity where in general CNTs grown by arc-discharge are close to perfectly crystalline, while CVD-grown CNTs have large defect densities. Mechanical deformation also affects these properties, even without introducing defects. When CNTs are bent they behave similarly to drinking straws, i.e. they buckle or ripple and their bending stiffness drops abruptly. In this thesis, the mechanical behaviour of individual CNTs and vertically aligned carbon nanofibers (VACNFs) has been studied by performing force measurements inside electron microscopes. Cantilevered CNTs, and VACNFs, were bent using a force sensor, yielding force-deflection curves while their structure was imaged simultaneously. We have found that CNTs grown by arc-discharge have a high enough crystallinity to possess a Young’s modulus close to the ideal value of 1 TPa. CVD-grown CNTs possess a Young’s modulus that is about one order of magnitude smaller, due to their large defect density. The VACNFs are yet another order of magnitude softer as a result of their cup-stacked internal structure.  We also found that a high defect density will increase the critical strain for the rippling onset and the relative post-rippling stiffness. Multi-walled CNTs with a small inner diameter are less prone to ripple and have a larger relative post-rippling stiffness. Our findings show large variations in the onset of rippling and the bending stiffness before and after rippling. These variations open up possibilities of tailoring the mechanical properties for specific applications. / Baksidetext Carbon nanotubes (CNTs) have extraordinary mechanical and electrical properties. Together with their small dimensions and low density, they are attractive candidates for building blocks in nanoelectromechanical systems (NEMS), and many other applications.  In this thesis the mechanical behaviour of individual CNTs and vertically aligned carbon nanofibers has been studied by performing force measurements inside electron microscopes. We have found that the mechanical behaviour is very sensitive to the defect density and the internal structure of the CNTs. The extraordinary properties are only attained by defect free CNTs and quickly deteriorate if defects are introduced to the structure. Mechanical deformations also alter these properties. Single-walled CNTs behave similarly to drinking straws when bent, i.e. they buckle, while the inner tubes of multi-walled CNTs prevent buckling. Instead a more distributed rippling pattern is created for multi-walled CNTs. Both these deformation behaviours will cause an abrupt drop in the bending stiffness, which is detrimental for many applications. The findings in this work will have implications for the design of future NEMS. / <p>Artikel 2 Image formation mechanisms tidigare som manuskript, nu publicerad: urn:nbn:se:kau:diva-16425 (MÅ 150924)</p>

Elastic properties of carbon nanotubes (CNT), boron nitride nanotubes (BNNT), double stranded DNA (dsDNA), paranemic-juxtapose crossover (PX-JX) DNA and dendrimer bound DNA are discussed in this thesis. Structural phase transitions of nucleic acids induced by external force, carbon nanotubes and graphene substrate are also studied extensively. Electrostatic interactions have a strong effect on the elastic properties of BNNTs due to large partial atomic charges on boron and nitrogen atoms. We have computed Young’s modulus (Y ) and shear modulus (G) of BNNT and CNT as a function of the nanotube radius and partial atomic charges on boron and nitrogen atoms using molecular mechanics calculation. Our calculation shows that Young’s modulus of BNNTs increases with increase in magnitude of the partial atomic charges on B and N atoms and can be larger than the Young’s modulus of CNTs of same radius. Shear modulus, on the other hand depends weakly on the magnitude of partial atomic charges and is always less than the shear modulus of the CNT. The values obtained for Young’s modulus and shear modulus are in excellent agreement with the available experimental results. We also study the elasticity of dsDNA using equilibrium fluctuation methods as well as nonequilibrium stretching simulations. The results obtained from both methods quantitatively agree with each other. The end-to-end length distribution P(ρ) and angle distribution P(θ) of the dsDNA has a Gaussian form which gives stretch modulus (γ1) to be 708 pN and persistence length (Lp) to be 42 nm, respectively. When dsDNA is stretched along its helix axis, it undergoes a large conformational change and elongates about 1.7 times its initial contour length at a critical force. Applying a force perpendicular to the DNA helix axis, dsDNA gets unzipped and separated into two single-stranded DNA (ssDNA). DNA unzipping is a fundamental process in DNA replication. As the force at one end of the DNA is increased the DNA starts melting above a critical force depending on the pulling direction. The critical force fm , at which dsDNA melts completely decreases as the temperature of the system is increased. The melting force in the case of unzipping is smaller compared to the melting force when the dsDNA is pulled along the helical axis. In the case of melting through unzipping, the double-strand separation has jumps which correspond to the different energy minima arising due to sequence of different base-pairs. Similar force-extension curve has also been observed when crossover DNA molecules are stretched along the helix axis. In the presence of mono-valent Na+ counterions, we find that the stretch modulus (γ1 ) of the paranemic crossover (PX) and its topoisomer juxtapose (JX) DNA structure is significantly higher (30 %) compared to normal B-DNA of the same sequence and length. When the DNA motif is surrounded by a solvent of divalent Mg2+ counterions, we find an enhanced rigidity compared to in Na+ environment due to the electrostatic screening effects arising from the divalent nature of Mg2+ counterions. This is the first direct determination of the mechanical strength of these crossover motifs which can be useful for the design of suitable DNA motifs for DNA based nanostructures and nanomechanical devices with improved structural rigidity. Negatively charged DNA can be compacted by positively charged dendrimer and the degree of compaction is a delicate balance between the strength of the electrostatic interaction and the elasticity of DNA. When the dsDNA is compacted by dendrimer, the stretch modulus, γ1 and persistence length, Lp decreases dramatically due to backbone charge neutralization of dsDNA by dendrimer. We also study the effect of CNT and graphene substrate on the elastic as well as adsorption properties of small interfering RNA (siRNA) and dsDNA. Our results show that siRNA strongly binds to CNT and graphene surface via unzipping its base-pairs and the propensity of unzipping increases with the increase in the diameter of the CNTs and is maximum on graphene. The unzipping and subsequent wrapping events are initiated and driven by van der Waals interactions between the aromatic rings of siRNA nucleobases and the CNT/graphene surface. However, dsDNA of the same sequence undergoes much less unzipping and wrapping on the CNT/graphene due to smaller interaction energy of thymidine of dsDNA with the CNT/graphene compared to that of uridine of siRNA. Unzipping probability distributions fitted to single exponential function give unzipping time (τ) of the order of few nanoseconds which decrease exponentially with temperature. From the temperature variation of unzipping time we estimate the free energy barrier to unzipping. We have also investigated the binding of siRNA to CNT by translocating siRNA inside CNT and find that siRNA spontaneously translocates inside CNT of various diameters and chiralities. Free en- ergy profiles show that siRNA gains free energy while translocating inside CNT and the barrier for siRNA exit from CNT ranges from 40 to 110 kcal/mol depending on CNT chirality and salt concentration. The translocation time τ decreases with the increase of CNT diameter having a critical diameter of 24 A for the translocation. After the optimal binding of siRNA to CNT/graphene, the complex is very stable which can serve as siRNA delivery agent for biomedical applications. Since siRNA has to undergo unwinding process in the presence of RNA-induced silencing complex, our proposed delivery mechanism by single wall CNT possesses potential advantages in achieving RNA interference (RNAi).

Downscaling of the contact length Lc of a side-contacted carbon nanotube field-effect transistor (CNTFET) is challenging because of the rapidly increasing contact resistance as Lc falls below 20–50 nm. If in agreement with existing experimental results, theoretical work might answer the question, which metals yield the lowest CNT–metal contact resistance and what physical mechanisms govern the geometry dependence of the contact resistance. However, at the scale of 10 nm, parameter-free models of electron transport become computationally prohibitively expensive. In our work we used a dedicated combination of the Green function formalism and density functional theory to perform an overall ab initio simulation of extended CNT–metal contacts of an arbitrary length (including infinite), a previously not achievable level of simulations. We provide a systematic and comprehensive discussion of metal–CNT contact properties as a function of the metal type and the contact length. We have found and been able to explain very uncommon relations between chemical, physical and electrical properties observed in CNT–metal contacts. The calculated electrical characteristics are in reasonable quantitative agreement and exhibit similar trends as the latest experimental data in terms of: (i) contact resistance for Lc = ∞, (ii) scaling of contact resistance Rc(Lc); (iii) metal-defined polarity of a CNTFET. Our results can guide technology development and contact material selection for downscaling the length of side-contacts below 10 nm.

Machine Learning interatomic potentials (ML-IAP) are currently the most promising Non-empirical IAPs for molecular dynamic (MD) simulations. They use Machine Learning (ML) methods to fit the potential energy surface (PES) with large reference datasets of the atomic configurations and their corresponding properties. Promising near quantum mechanical accuracy while being orders of magnitudes faster than first principle methods, ML-IAPs are the new “hot topic” in material science research. Unfortunately, most of the available publications require advanced knowledge about ML methods and IAPs, making them hard to understand for beginners and outsiders. This work serves as a plain introduction, providing all the required knowledge about IAPs, ML, and ML-IAPs from the beginning and giving an overview of the most relevant approaches and concepts for building those potentials. Exemplary a gaussian approximation potential (GAP) for amorphous carbon is used to simulate the defect induced deformation of carbon nanotubes. Comparing the results with published density-functional tight-binding (DFTB) results and own Empirical IAP MD-simulations shows that publicly available ML-IAP can already be used for simulation, being indeed faster than and nearly as accurate as first-principle methods. For the future two main challenges appear: First, the availability of ML-IAPs needs to be improved so that they can be easily used in the established MD codes just as the Empirical IAPs. Second, an accurate characterization of the bonds represented in the reference dataset is needed to assure that a potential is suitable for a special application, otherwise making it a 'black-box' method.:1 Introduction 2 Molecular Dynamics 2.1 Introduction to Molecular Dynamics 2.2 Interatomic Potentials 2.2.1 Development of PES 3 Machine Learning Methods 3.1 Types of Machine Learning 3.2 Building Machine Learning Models 3.2.1 Preprocessing 3.2.2 Learning 3.2.3 Evaluation 3.2.4 Prediction 4 Machine Learning for Molecular Dynamics Simulation 4.1 Definition 4.2 Machine Learning Potentials 4.2.1 Neural Network Potentials 4.2.2 Gaussian Approximation Potential 4.2.3 Spectral Neighbor Analysis Potential 4.2.4 Moment Tensor Potentials 4.3 Comparison of Machine Learning Potentials 4.4 Machine Learning Concepts 4.4.1 On the fly 4.4.2 De novo Exploration 4.4.3 PES-Learn 5 Simulation of defect induced deformation of CNTs 5.1 Methodology 5.2 Results and Discussion 6 Conclusion and Outlook 6.1 Conclusion 6.2 Outlook

Kohlenstoffnanoröhren (engl. carbon nanotubes) bieten hervorragende elektrische Eigenschaften für neuartige Feldeffekttransistoren (engl. field-effect transistors) auf engstem Raum. Eine Möglichkeit zur Verbesserung der elektrischen Eigenschaften bietet eine geeignete Passivierung mit Hexamethyldisiloxan. In dieser Arbeit werden eine Flüssig- und eine Gasphasenbehandlung von Siliziumoxid-Oberflächen mit Hexamethyldisiloxan untersucht. Die Oberflächen werden dabei in wenigen Minuten hydrophobiert. Nach längeren Behandlungszeiten werden Wasserkontaktwinkel von 95° erreicht, die auch noch nach mehreren Tagen und einer Woche nachweisbar sind. In der Anwendung auf Kohlenstoffnanoröhren-Feldeffekttransistoren (engl. carbon nanotube field-effect transistors) wird die Hysterese um durchschnittlich 30 % gesenkt. Das Ziel der Behandlung wurde damit erreicht und lässt sich auf die erfolgreiche Beseitigung von Ladungsfallen durch Adsorbate zurückführen. Zusätzlich sinkt der An-Strom um 60 %. Für gute An-Aus-Verhältnisse über mehrere Größenordnungen bedeutet das jedoch keine drastische Verschlechterung der Schalteigenschaften. Die in dieser Arbeit vorgeschlagene Hexamethyldisiloxan-Gasphasenbehandlung kann daher erfolgreich zur Verringerung der Hysterese in Kohlenstoffnanoröhren-Feldeffekttransistoren eingesetzt werden.:1 Einleitung 6 2 Material und Methoden 8 2.1 Siliziumoxid-Oberflächen 8 2.2 Hexamethyldisiloxan als Passivierungsmittel 8 2.3 Flüssigphasenbehandlung 9 2.4 Gasphasenbehandlung 10 2.5 Kontaktwinkelmessung 11 3 Feldeffekttransistoren auf der Basis von Kohlenstoffnanoröhren 13 3.1 Kohlenstoffnanoröhren 13 3.1.1 Struktur und Nomenklatur 13 3.1.2 Elektrische Eigenschaften 14 3.2 Kohlenstoffnanoröhren-Feldeffekttransistoren 16 3.2.1 Aufbau, Herstellung und Funktionsprinzip 16 3.2.2 Kenngrößen zur FET-Charakterisierung 17 3.2.3 IU-Messung 19 4 Ergebnisse und Auswertung 20 4.1 Ergebnisse der Kontaktwinkelmessungen 20 4.1.1 Referenzmessungen auf Siliziumoxidoberflächen 20 4.1.2 Flüssigphasenbehandlung 21 4.1.3 Gasphasenbehandlung 21 4.1.4 Fehlerbetrachtung 25 4.1.5 Vergleich von Flüssig- und Gasphasenbehandlung 26 4.2 HMDSO-Behandlung von CNTFETs 27 4.2.1 Ergebnisse der IU-Messungen 28 4.2.2 Fehlerbetrachtung 31 5 Zusammenfassung der Ergebnisse und Ausblick 32

No description available.

No description available.

The carbon nanotube-based field effect transistor (CNTFET) possesses the potential to overcome limitations of state-of-the-art technologies such as silicon-based complementary metal-oxide-semiconductors. However, the carbon nanotube (CNT) technology is still at its infancy and technology development is still necessary to exploit the CNT properties such as high charge carrier mobility, high current carrying capability, one-dimensional charge transport and their versatile integrability. Within this work significant progress has been achieved scientifically and technologically in the advance of high frequency (HF) CNTFETs for analog applications. According to simulations by others, a technology flow has been developed based on electron beam lithography for bottom gated HF CNTFETs which outperform state-of the art top gate architectures with respect to their parasitic capacitances. Moreover, the impact of electrostatic doping on the CNTFETs has been investigated. In particular, the dynamics of water desorption from the CNTFETs and the related reduction of p-type doping was investigated and the different impact of the n-type dopant polyethylenimine onto the channel region and contact region could be separated for the first time. Furthermore, the impact of doped CNT bundles on the device performance has been studied. It could be shown in detail for the first time, that high off-state source-drain leakage currents can be due to bundled semiconducting CNTs and does not necessarily imply the presence of metallic CNTs. The within the framework of this thesis designed and realized HF CNTFETs are operating in the GHz range with cut-off frequencies up to 14 GHz and maximum frequencies of oscillation up to 6 GHz at a channel length of 280 nm. Moreover, the impact of the spacer between the source-/ drain- to the gate electrode on the HF properties of the CNTFETs has been investigated experimentally for the first time. Simulations by others have successfully confirmed that a symmetrical reduction of the source to gate electrode spacer results in an increased device speed. By asymmetrically reducing the source to gate electrode spacer and in parallel increasing the drain-to-gate electrode spacer the device speed can be further enhanced. Moreover, within this work it has been experimentally indicated for the first time that the device properties of HF CNTFETs can be tuned by different device geometries towards either highest linearity or speed.

No description available.