Interest in nanoscale heat and mass transport has been augmented through current trends in nanotechnology research. The theme of this dissertation is to characterize electric charge, mass and thermal transport at the nanoscale using a fundamental molecular simulation method, namely molecular dynamics. This dissertation reports simulations of (1) ion intake by carbon nanotubes, (2) hydrogen storage in carbon nanotubes, (3) carbon nanotube growth and (4) nanoscale heat transfer. Ion transport is investigated in the context of desalination of a polar solution using charged carbon nanotubes. Simulations demonstrate that when either a spatially or temporally alternating charge distribution is applied, ion intake into the nanotubes is minimal. Thus, the charge distribution can either be maintained constant (for ion encapsulation) or varied (for water intake) in order to achieve different effects. Next, as an application of mass transport, the hydrogen storage characteristics of carbon nanotubes under modified conditions is reported. The findings presented in this dissertation suggest a significant increment in storage in the presence of alkali metals. The dependence of storage on the external thermodynamic conditions is analyzed and the optimal range of operating conditions is identified. Another application of mass transport is the growth mode of carbon nanostructures (viz. tip growth and base growth). A correct prediction of the dominant growth mode depends on the energy gain due to the addition of C-atoms from the carbon-metal catalyst solution to the graphene sheets forming the carbon nanostructures. This energy gain is evaluated through molecular dynamics simulations. The results suggest tip growth for Ni and base growth for Fe catalysts. Finally, unsteady nanoscale thermal transport at solid-fluid interfaces is simulated using non-equilibrium molecular dynamics simulations. It is found that the simulated temperature evolution deviates from an analytical continuum solution due to the overall system heterogeneity. Temperature discontinuities are observed between the solid-like interfaces and their neighboring fluid molecules. With an increase in the temperature of the solid wall the interfacial thermal resistance decreases. / Ph. D.

Carbon nanotubes (CNTs) are known for their exceptional electrical, thermal, mechanical, optical, and chemical properties. With the significant progress in recent years on synthesis, purification and integration challenges, CNT network/array based thin-film transistors (TFTs) are likely to play a critical role as the building blocks of future electronics. CNT-TFTs can find applications in flexible, transparent and energy-efficient circuits, e-displays, solar cells, RFID tags, e-paper, touch screens, implantable medical devices and chemical/bio/optical sensors. CNTs in CNT-TFTs are deposited on low thermal conductivity substrates which can impede the heat dissipation resulting in high temperature. The excessive self-heating in CNT-TFTs can degrade the electrical and thermal performance and could potentially lead to failure of the devices. Therefore, the issues related to operational reliability of CNT-TFTs arising from the self-heating effects need to be examined and studied. In the present work, a computational approach is developed and employed to study the electrical and thermal transport in CNT-TFTs. The modeling framework can predict the current and temperature profile of CNT network/array and the supporting structure. The model is validated against the experimental results. In case of CNT network TFTs, the computational method allows us to examine the role of various device parameters such as network morphology (i.e., network density, CNT junction topology, and CNT length and alignment distribution) and channel geometry (i.e., channel length and width) on heat dissipation and thermal reliability. The simulation results help interpret experimental data and provide the quantitative information about the thermal boundary conductances at CNT junctions and CNT-substrate interfaces in CNT-TFTs. The findings suggest that the structure of CNT junctions on substrate can become very critical in CNT network TFTs as the lack of contact with the substrate at these junctions can lead to junction temperatures hundreds of degrees higher than the rest of the device, which will severely deteriorate the performance of these devices. High-field breakdown study of CNT network TFTs is also conducted which provides guidelines for the design and optimization with respect to aforementioned parameters in order to enhance the performance and reliability. Dense CNT arrays are preferred for better electrical performance in CNT array TFTs, but they also experience electrostatic and thermal cross-talk which can adversely affect the device performance. These effects have been studied in details. The role of trap charges in CNT array TFTs is also investigated to understand and mitigate hysteresis. Lastly, CNT-liquid crystal composites are studied using dissipative particle dynamics (DPD) technique with the aim to understand how the CNT concentration in composite affects the alignment of liquid crystals and to explore the method of CNT alignment using liquid crystals.

The continued miniaturization and proliferation of electronics is met with significant thermal management challenges. Decreased size, increased power densities, and diverse operating environments challenge the limitations of conventional thermal management schemes and materials. To enable the continuation of these trends thermal interface materials (TIMs) that are used to enhance heat conduction and provide stress relief between adjacent layers in a electronic package must be improved. Forests comprised of nominally vertically aligned carbon nanotubes (CNTs), having outstanding thermal and mechanical properties, are excellent candidates for next-generation thermal interface materials (TIMs). However, despite nearly a decade of research, TIMs based on vertically aligned CNT forests have yet to harness effectively the high thermal conductivity of individual CNTs. One of the key obstacles that has limited the performance of CNT TIMs is the presence of high thermal contact resistances between the CNT free ends and the surfaces comprising the interface. The aim of this research is to better understand the mechanisms by which the thermal contact resistance of CNT forest thermal interfaces can be reduced and to use this understanding towards the design of effective and to scalable processing methods. Contact area and weak bonding between the CNT tips and opposing surface are identified as factors that contribute significantly to the thermal contact resistance. Three strategies are explored that utilize these mechanisms as instruments for reducing the contact resistance; i) liquid softening, ii) bonding with surface modifiers, and iii) bonding with nanoscale polymer coatings. All three strategies are found to reduce the thermal contact resistance at the CNT forest tips to below 1 mm2-K/W, a value to where it is no longer the factor limiting heat conduction in CNT forest TIMs. These strategies are also relatively low-cost and amenable to scaling for production when compared to existing metal-based bonding strategies.

A preliminary gas detection study was conducted on as-grown multi-walled carbon nanotubes and anodized aluminum oxide (MWNTs/AAO) template. The material demonstrated room temperature gas sensitivity and p-type semiconductor characteristics. Plasma-etched MWNTs/AAO templates were employed to construct capacitive gas sensors. The capacitances of the sensors were sensitive to both reducing and oxidizing gases at room temperature. Single-walled carbon nanotubes (SWNTs) dispersed in binder andamp;aacute;-terpineol were applied on sensor platforms to form resistive gas sensors. The sensors demonstrated excellent sensitivity to low concentrations of reducing and oxidizing gases at room temperature, which suggests the p-type semiconducting behavior of SWNTs. The sensor recovery was found to be incomplete at room temperature in flow of nitrogen and air, thus possible solutions were investigated to enhance sensor performance. The sensor operating principles and suggestions for possible future work are discussed. The room temperature and air background functionality of the sensor suggest that SWNT is a promising gas sensing material for application in ambient conditions.

As electronics reach nanometer size scales, new avenues of integrating biology and electronics become available. For example, nanoscale field-effect transistors have been integrated with single neurons to detect neural activity. Researchers have also used nanoscale materials to build electronic ears and noses. Another exciting development is the use of nanoscale biosensors for the point-of-care detection of disease biomarkers. This thesis addresses many issues that are relevant for electrical sensing applications in biological environments. As an experimental platform we have used carbon nanotube field-effect transistors for the detection of biological proteins. Using this experimental platform we have probed many of properties that control sensor function, such as surface potentials, the response of field effect transistors to absorbed material, and the mass transport of proteins. Field effect transistor biosensors are a topic of active research, and were first demonstrated in 1962. Despite decades of research, the mass transport of proteins onto a sensor surface has not been quantified experimentally, and theoretical modeling has not been reconciled with some notable experiments. Protein transport is an important issue because signals from low analyte concentrations can take hours to develop. Guided by mass transport modeling we modified our sensors to demonstrate a 2.5 fold improvement in sensor response time. It is easy to imagine a 25 fold improvement in sensor response time using more advanced existing fabrication techniques. This improvement would allow for the detection of low concentrations of analyte on the order of minutes instead of hours, and will open the door point-of-care biosensors. / Graduation date: 2011

This dissertation presents research on scanning probe microscopy and spectroscopy of graphene and carbon nanotubes. In total three experiments will be discussed. The first experiment uses a scanning tunneling microscope (STM) to study the topographic and spectroscopic properties of graphene on hexagonal boron nitride (hBN). Graphene was first isolated and identified on SiO₂ substrates, which was later found to be the source of graphene quality degradation, e.g. large surface roughness, increased resistivity and random doping etc. Researchers have been trying to replace SiO₂ with other materials and hBN is by far the most successful one. Our STM study shows an order of magnitude reduction in surface roughness and electrostatic potential variation compared with graphene on SiO₂.The second experiment shows a novel quantum interference effect of electron waves in graphene, loosely referred to as "Friedel oscillations." These arise when incident electron waves interfere with waves scattered from defects in the sample. This interference pattern shows up as a spatial variation in the local density of states, which can be probed by the STM. We measured such Friedel oscillations in graphene near step edges of hBN. Due to its peculiar band structure, the oscillations in graphene have a faster decay rate and their wavelength is an order of magnitude longer than similar oscillations previously observed on noble metal surfaces. By measuring the dependence of the Friedel oscillations on electron energy, we map out the band structure of graphene. The last experiment studies a different system: carbon nanotube quantum dots. By combining scanning probe microscopy and transport measurements, we obtain spatial information about quantum dots formed in a carbon nanotube field effect transistor. We also demonstrate the ability to tune the coupling strength between two quantum dots in series.

Carbon nanotubes are intriguing materials and extensively studied for both their fundamental properties and extraordinary performance in various applications during the last 20 years. They are extremely small in diameter, light in weight, sensitive to the environment, strong, and chemically stable. They can be either metallic or semiconducting depending on their species. The semiconducting species can absorb and emit light in a wide range of wavelengths. These outstanding properties of carbon nanotubes promise abundant applications that may be revolutionary. The opto-electronic behaviour of a single-walled carbon nanotube (SWCNT) is extremely sensitive to its physical structure and ambient environment. Structural defects and surrounding environment are extrinsic influential factors that often obscure the understanding of the intrinsic behaviour. Progress on SWCNT synthesis has been made continuously but not until the last 10 years, have single SWCNTs been isolated individually and from substrates so that their fluorescence can be detected. The fundamental science of an optically generated exciton (an electron-hole pair) in an ideal semiconducting SWCNT is not fully understood despite many studies of exciton behaviour using various optical approaches. The major challenge is controlling SWCNT sample qualities. SWCNT's fundamental properties, such as the absorption cross section, quantum efficiency, radiative and nonradiative lifetimes, remain under debate. Knowing the intrinsic SWCNT properties is essential to understand exciton transport and relaxation mechanisms. To minimize the extrinsic effects, we have selected high-quality unprocessed SWCNTs for investigation. Collaboration with Dr. P. Finnie and Dr. J. Lefebvre at National Research Council Canada, allow us to access pristine SWCNTs individually. Since the emission from a single SWCNT is low, it requires unconventional methods to measure the PL dynamics. Suggested by the results, exciton transport in a semiconducting SWCNT is diffusional at room temperature, with high diffusivity (130 -350 cm^2/s) and long diffusion length (1 - 5 µm). At lower temperatures, we observed a more efficient exciton-exciton interaction that suggests the contribution from hot excitons or a longer existence of delocalized excitons. Highly efficient exciton-exciton annihilation and long coherence time in a SWCNT are promising for making a single-photon source at near-infrared wavelength range and developing quantum computers. / Thesis (Ph.D, Physics, Engineering Physics and Astronomy) -- Queen's University, 2013-12-06 09:52:51.136

Since their discovery Carbon Nanotubes (CNTs) have attracted much interest from many fields of the scientific community owing to their range of unique and impressive properties. Measurements of the mechanical properties of these nanoscale molecules have shown strengths up to five times greater than that of steel at only a quarter of the density. Consequently many have attempted to unlock these remarkable properties by creating nano-composite structures where CNTs effectively reinforce materials with little increase in density. Unfortunately the tendency of CNTs to form agglomerations when allowed to disperse in fluid suspensions has made this process non trivial, and led to difficulties in achieving effective reinforcement when simply mixing CNTs into a matrix material. As a result it has become clear that new approaches to composite construction will be required if effective composite reinforcement using CNTs is to be achieved. Recent advances in CNT synthesis using Chemical Vapour Deposition (CVD) where tall forests of these nanoparticles are grown from the vapour phase have begun to solve the agglomeration problem. These forests are produced in aligned and dispersed arrays, and wetting of these structures with polymer matrices has demonstrated improvements in modulus of several hundred percent. These improvements arise as the CNTs retain both the dispersion and alignment of the forest when incorporated into the matrix thus overcoming the difficulties observed using traditional manufacture methods. New complications arise when attempting to extend these promising results to larger scale composite components owing to the typically millimetre size of CVD grown vertically aligned CNT (VACNT) forests. From these results it follows that to create large composite parts it will be required to incorporate many individually CVD grown VACNT forests into a single composite structure. Strategies to achieve such a composite are being developed, with a range of ideas extending from knowledge gained from the emerging technology of additive manufacture (AM) described as ‘...the process of joining materials to make objects from 3D model data, usually layer upon layer....’. Indeed it is desirable to reinforce materials used in AM processes and the nano scale diameter of CNTs makes them the perfect choice owing to their high aspect ratios at the micron scale. In this thesis investigations are conducted into the feasibility of manufacturing CNT composite structures using CVD grown forests and AM techniques. These investigations include measurement of the anisotropic mechanical properties of composite samples, and studies of the wetting interactions that occur between CNT forests and polymer materials. Composite samples are constructed and tested mechanically in the transverse orientation and results compared to traditional fibre composite reinforcement models in order to understand the material properties that can be expected if such an AM process is achieved. Results show greater mechanical improvements in transverse modulus than expected, and these results are attributed to the wavy nature of individual CNTs within forest structures providing multi directional reinforcement to the matrix material. Further studies are conducted to investigate the flow of molten thermoplastic materials into CNT forest structures under capillary driven flow. Thermoplastics were allowed to flow into VACNT forests before being cooled and inspected using micro x-ray computed topography (μ-CT) to gain an understanding of the wetting mechanism. Results from μ-CT scans show that the polymer flows into the structure in peaks of similar radius. Finally dynamic investigations were conducted into the fast capillary driven flow of a low viscosity thermoset resin into VACNT forests using a high speed camera. Results are fitted to traditional models for dynamic capillary driven flow in porous media and an effective radius and porosity is calculated for VACNT forests. Experimental values illustrate that these nanoscale structures still fit to traditional flow models of fluids where the height of capillary rise is proportional to the square of the elapsed time. These results provide a further step in understanding methods of incorporating many VACNT structures into polymeric matrices to achieve large scale effective polymer VACNT composite materials.

Thermoelectric generators are solid state machines used to convert temperature gradients into electrical energy. They are formed by several thermoelectric units connected electrically in series and thermally in parallel. These units are made by creating a junction between a p-type and an n-type conductor. This investigation documents the characterisation of the thermoelectric properties of carbon nanotube (CNT) films and the fabrication process of carbon nanotube-based thermoelectric devices. The Seebeck coefficient is a intrinsic property of a thermoelectric material that correlates the voltage produced by a conductor and the temperature gradient applied to it. To measure the Seebeck coefficient of films, an experimental set-up was fabricated and calibrated using constantan as standard material. CNT films of aligned nanotubes fabricated using a chemical vapour deposition method were analysed. The Seebeck coefficient along and across the samples did not show significant variations, with values between 40$\mu$V/K and 80$\mu$V/K. Using these CNT films, thermoelectric cells were fabricated with the CNT as the p-type conductor and constantan as the n-type. As a proof of concept, two hand-made thermoelectric generators were assembled by connecting hundreds of these thermoelectric cells. These devices were subjected to a temperature gradient of $\approx$200K, which was enough to produce enough power to light an LED. Other analytical techniques were used to characterise the materials used in this work. Electrical conductivity measurements, thermogravimetric analysis, Raman spectroscopy and scanning electron microscopy were performed. Using a deposition technique, films of nanotubes were produced from a liquid phase. The impact of the production method on their properties was evaluated. Characterisation equipment was developed to measure the Seebeck coefficient and thermal conductivity. Thermoelectric devices made with the carbon nanotube films were fabricated and characterised. The values of thermal conductivity of the CNT films analysed in this work are between 0.86Wm$^{-1}$K$^{-1}$. The electrical conductivity of these materials is between 3500Sm$^{-1}$ and 14100Sm$^{-1}$. The maximum figure of merit of the carbon nanotube thermoelectric devices fabricated in this work is $ZT$=0.35.

The main objectives of the present Ph.D. thesis are comprehensive studies on activity, selectivity and stability of iron catalysts supported on carbon nanotubes (CNTs) for Fischer-Tropsch (FT) reactions. In order to prepare iron catalyst supported on CNTs, it was necessary to study CNT synthesis in bulk scale. Therefore, a part of this research was devoted to the production and characterization of CNTs. High purity, aligned films of multi-walled carbon nanotubes were grown on quartz substrates by feeding a solution of ferrocene in toluene, in a carrier gas of Ar/H2, into a horizontal chemical vapour deposition (CVD) reactor. Results for CNTs synthesized using a wide range of toluene concentrations indicated that, for carbon concentrations higher than ~9.6 mol/m3, catalyst deactivation occurs due to encapsulation of iron metal particles.<p> As the first step of catalyst development for FT reactions a fixed bed micro-reactor system was built and the effects of acid treatment on the activity, product selectivity and stability of iron Fischer-Tropsch catalysts supported on carbon nanotubes were studied. The results of Raman analysis showed that the acid treatment increased the number of functional groups as anchoring sites for metal particles. Fe catalysts supported on CNTs which were pre-treated with nitric acid at 110°C were more stable and active compared to the un-treated catalysts. In order to study the effects of catalytic metal site position on FT reactions, a method was developed to control the position of the deposited metal clusters on either the inner or outer surfaces of the CNTs. According to the results of the FT experiments, the catalyst with catalytic metal sites inside the pores exhibited higher selectivity (C₅⁺ = 36 wt%) to heavier hydrocarbons compared to one with sites on the outer surfaces (C₅⁺ = 24 wt%) . In addition, deposition of catalytic sites on the interior surfaces of the nanotubes resulted in a more stable catalyst.<p> The effects of pore diameter and structure of iron catalysts supported on CNTs on Fischer-Tropsch reaction rates and selectivities were also studied. In order to examine the effects of pore diameter, two types of CNTs with similar surface areas and different average pore sizes (12 and 63 nm) were prepared. It was found that the deposition of metal particles on the CNT with narrow pore size (in the range of larger than 10-15 nm) resulted in more active and selective catalyst due to higher degree of reduction and higher metal dispersion.<p> Promotion of the iron catalyst supported on CNTs with Molybdinium in the range of 0.5-1 wt % resulted in a more stable catalyst. Mo improves the stability of the iron catalyst by preventing the metal site agglomeration. Promotion of the iron catalysts with potassium increased the activity of FT and water-gas-shift reactions and the average molecular weight of the hydrocarbon products. Promotion of the iron catalyst supported on CNTs with 0.5% Cu and 1wt% K resulted in an active (5.6 mg HC/g-Fe.h), stable and selective catalyst (C₅⁺ selectivity of 76%) which exhibited higher activity and better selectivity compared to the similar catalysts reported in the literature. Kinetic studies were conducted to evaluate reaction rate parameters using the developed potassium and copper promoted catalyst. It was found that the CO₂ inhibition is not significant for FT reactions. On the other hand, water effects and presence of vacant sites should be considered in the kinetic models. A first-order reaction model verified that the iron catalyst supported on CNTs is more active than precipitated and commercial catalysts. The results of the present Ph.D. thesis research provide a map for designing catalysts using carbon nanotubes as a support. The key messages of the present thesis are as follows:<p> 1- If the interaction of the metal site and support is strong, which poses negative effects on the catalytic performance, carbon nanotubes can be one solution.<p> 2- Acid pre-treatments are required prior to impregnating nanotubes with metal salt solution. Also, the strong acid treatment should be used for deposition of catalytic sites inside the pores of nanotubes.<p> 3- The structure and pore size of nanotubes have significant influence on the stability, activity and selectivity of the target catalyst.<p> 4- The position of the catalytic sites has to be selected based on the type of reaction. In the case of Fischer-Tropsch reactions, the deposition of catalytic sites inside the pores of nanotubes results in higher activity, longer life span.<p> The outcome of this Ph.D. thesis has been published/submitted in the form of 13 journal papers, one patent, one technical report and presented at 11 conferences.