Classical molecular dynamics (MD) simulations have been widely used to study the physical properties of nanomaterials. However, MD suffers from the well-known drawback where it cannot, in the absence of special high-performance computing environments, simulate processes that take longer than microseconds. Nevertheless, many important processes in various fields of science and engineering take place on time scales that cannot be reached by MD. For example, while MD simulations have been used extensively to study the plasticity of nanostructures such as proteins and nanowires, their time scale limitations have prevented the study of the deformation mechanisms at experimentally-relevant forces, time scales and strain rates. In this thesis, a generic history-penalized self-learning metabasin escape (SLME) algorithm, which efficiently explores the potential energy surface, is utilized to overcome this issue by studying three canonical problems of interest: the unfolding of biological proteins under experimentally-relevant mechanical forces; the strain-rate-dependent plastic deformation mechanisms of bicrystalline FCC metal nanowires, and finally the constant stress (creep-induced) plastic deformation of single crystalline metallic nanowires at experimental time scales. The SLME method is first utilized to study the force-induced unfolding of biological proteins. The long time scale simulations not only provide atomistic details of time-dependent structural evolution under experimentally-relevant forces, but also show novel intermediate states that cannot be observed in MD simulations, which sheds new insight into the understanding of protein unfolding dynamics. This method is then utilized to investigate the strain rate effect on the plastic mechanisms of bicrystalline metal nanowires. A strain-rate-dependent incipient plasticity and yielding transition for bicrystalline metal nanowires at experimentally-relevant strain rates is observed. This transition leads to a ductile-to-brittle transition in failure mode, which is driven by differences in dislocation activity and grain boundary mobility at low strain rate as compared to the high strain rate case. Finally, the SLME method is also utilized to study the creep behavior of single crystalline metal nanowires at experimental time scales. Both copper and silver nanowires show significantly increased ductility and superplasticity under experimentally-relevant creep stresses, where the superplasticity is driven by a thermally-activated transition in defect nucleation from twinning to trailing partial dislocations at the micro or millisecond timescale.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 36). / Forecasting the future prices or returns of a security is extraordinarily difficult if not impossible. However, statistical analysis of a basket of highly correlated securities offering a cross-sectional representation of a particular sector can yield information that is potentially tradable. Securities related to the fossil fuels industry are used as the basis of a practical application to two distinct forecasting techniques. The first method, forecasting using conditional multivariate Gaussian statistics, was shown to yield, in a relative sense, the best results for those securities which exhibited a high correlation with the rest of the basket. For the second method, principal component analysis was done on a basket of commodity futures to reveal a small number of dominant factors governing the movements of the portfolio. Autoregressive models were then applied to both the factors and futures, but results showed both to be essentially Markov processes. / by Derek Z. Li. / S.B.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 309-328). / Part I: A classic problem in the field of fluid-structure interaction is the flapping-flag instability. Fluid-mechanical studies of the phenomenon date back to the 19th century, increased in number in recent years with increasingly accurate representations for the coupled fluid-structure interaction. The problem continues to attract attention because the effect of fluid forces and aspect ratio on stability is non-obvious. In the first part of the flapping studies, we examine three-dimensional effects on the flapping dynamics of a flag, modeled as a thin membrane, in a uniform fluid inflow. We consider periodic span-wise variations of length (ignoring edge effects) characterized by discrete span-wise wavenumber. Using linear stability analysis we show the increase in stability with discrete span-wise wavenumber. We confirm the stability analysis and study the nonlinear responses of three-dimensional flapping, using direct numerical simulation of the Navier-Stokes equations on a moving body-fitted computational grid for thin membrane structure undergoing arbitrarily (large) displacement. We perform direct numerical simulations, initialized using normal modes we derive, up to Reynolds number 1000 based on L. For nonlinear evolutions, we identify and characterize the effect of span-wise variations on the fundamental modes and responses of flapping in terms of span-wise standing wave (SW) and travelling wave (TW) modes respectively in the absence and presence of cross flow; and their corresponding flag displacements and wake vortex structures. We report for TW, the flag flapping and vortex shedding frequencies and angles are matched, and are related to the corresponding shedding frequency of SW. When both SW and TW modes are present due to stabilization of drag by the cross-flow, the fluid-flag response trends to be dominated over time by TW with continuous wake structure. In the second part of the flapping work we investigate the absolute or convective nature of the instability of a two-dimensional flapping filament submerged in a uniform fluid inflow. When the structure-to-fluid mass ratio is zero, we show that two families of flapping waves exist, with phase velocities that are equal in magnitude and have opposite signs, increasing the mass ratio for a given Reynolds number increases the phase velocity of the waves propagating in the same direction as the flow, and decreases the phase velocity of the waves propagating opposite to the flow. Using a linearized energy conservation law we show that after a critical value of mass ratio is exceeded the flapping instability is sustained when the fast (positive energy) and the slow (negative energy) waves coalesce creating waves with zero energy which do not require an energy source or a sink to be sustained, and grow exponentially in time. Under such conditions an analytical condition for absolute instability is derived. We further show based on a group velocity criterion, that when the two characteristic speeds have opposite signs the instability is absolute, where as if they have the same sign the instability is convective. A range of mass ratio regimes is found where the instability is absolute and where it is convective; with the unstable flapping amplitude at the instability threshold, satisfying the Klein-Gordon equation. / Part II: Accurate prediction of the highly mixed flow in the near field of a surface ship is a challenging and active research topic in Computational Ship Hydrodynamics. The disparity in length and time scales recognizes the importance of accurate bubble source and mixed-phase flow models; whereas the current state of the art models are adhoc at best. Second part of the thesis details the air entrainment characteristics in the incompressible highly variable density turbulent flow-field behind a canonical stern with the inclusion of simple speed/geometry/Reynolds number effects. Using high-resolution two-phase flow data sets generated from high fidelity simulations of a canonical stern simulated down to the scales of bubble entrainment. The study details key variables for: (i) characterization of wake structure, near-wake air entrainment and the nature of incompressible variable density turbulence, underlining the major implications and dominant terms by studying the dynamics of the continuity equation, the momentum equation, the density variance equation, the turbulent mass flux and the turbulent kinetic energy; (ii) the role of non-Boussinesq effects and turbulent mass flux in the wake of the stern, identifying the breaking event to be related to the air-entrainment and subsequent generation of turbulent mass flux and establishing the density intensity as an effective metric; (iii) develop and a priori validate novel multiphase models for turbulent mass flux and turbulent kinetic energy using gradient hypothesis and measuring the model performance for varied geometry/speed/Reynolds number effects. The first part of the thesis advances our understanding in varying applications ranging from the biomechanics of snoring, to improving novel designs for flow energy harvesters. The second part presents a methodology, using high-fidelity simulations coupled to physics-based parameterization of near-field air entrainment about surface ships to help improve mixed-phase turbulent flow models in Computational Ship Hydrodynamics. / by Sankha Banerjee. / Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 83-86). / As the workforce within manufacturing grows older, especially within aircraft manufacturing, the need for new technologies to assist workers arises. If a technology could offer improvements to an aircraft manufacturing laborer's efficiency, as well as reduce the load on his body, it could potentially see vast use. This thesis discusses a potential solution to these issues - the Supernumerary Robotic Limbs (SRL). These limbs could potentially increase the workspace of the human operator to him more efficient, as well as reduce the load on the human while he performs staining tasks. It accomplishes this by providing the worker with extra arms in the form of a wearable backpack. This thesis first evaluates how the torques imposed on a human are affected when he uses an SRL-like device to help bear a static load. It is shown that the human work load necessary to bear such a load is reduced substantially. The second focus of this thesis is the skill acquisition. A data-driven approach is taken to learn trajectories and a leader-follower coordination relationship. This is done by generating teaching data representing trajectories and coordination information with two humans, then transferring the pertinent information to a robot that assumes the role of the follower. This coordination is validated in a simple one-dimension example, and is implemented on a robot that coordinates with a human leader during a control-box wiring task. / by Clark Davenport. / S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2015. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 85-87). / In this thesis, the suitability of RFID as a platform for real time applications in the Internet of Things is investigated. Two sample applications of RFID technology identify throughput and latency as key figures of merit. An experiment to measure these parameters in an existing implementation is devised and executed. The sources of latency are identified with a high degree of accuracy. Recommendations for improvement of the existing implementation are guided by the latency measurements. / by Erick William Fuentes. / S.M.

Thesis (B.S.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1996. / Includes bibliographical references (leaf 132). / by John Muir Kumph. / B.S.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1996. / by Craig Weidner. / M.S.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2005. / Includes bibliographical references (p. 39 ). / The effect of probe tilt angle on the quality of Scanning Tunneling Microscopy (STM) measurements was explored. A small but consistent improvement in slope accuracy was documented lending some support to the effort to develop a new, five-axis STM capable of tilting in a controlled manner while scanning. The objective of such a machine would be to allow its probe to trace the sample's contour with greater accuracy than the currently available three-axis STM can. It is postulated that an STM with a probe that can change its roll and pitch in addition to its position along the traditional x, y, and z axes would be capable of reducing imaging errors produced as a result of geometric constraints, lateral electron discharge effects, and the tendency for the tip to bend during scanning due to electrostatic surface forces. In order to quantify the effects of incorporating probe tilt into the scanning process, a traditional, three-axis STM was manipulated in a way that allowed a standard sample grid to be imaged using a probe that was placed at seven different angles of tilt ranging from -13 to +13 degrees. Twenty-five different cavities in a standard STM scanning sample were scanned at these seven angles to determine notable trends and effects in the images produced. / (cont.) It was determined that for each degree of angle change in the tilt of the probe, the slopes of the cavity walls imaged improved by an amount of slope equal to approximately 0.001 nm/nm, which corresponds to 0.0093% less imaging error. This seemingly trivial improvement in wall slope is significant in light of the fact that the change in slope per degree of probe tilt is on the same order of magnitude as the slopes of the cavity walls measured by the STM. / by Jonathan B. Hopkins. / S.B.

Thesis (Sc. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2007. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Includes bibliographical references (leaves 248-255). / Transverse jets form a dominant group of flow fields arising in many applications of modern energy utilization, including propulsion and effluent dispersion. Furthermore, they form canonical examples where the flow field is dominated by large-scale and small-scale vortical structures, whose inter-related dynamics is a challenging subject in modern fluid mechanics. This study seeks a mechanistic understanding of the vortical structures of the transverse jet and their evolution. A set of massively parallel three-dimensional vortex simulations of high-momentum transverse jets at intermediate Reynolds number, utilizing a discrete filament representation of the vorticity field to capture stretching and tilting of vorticity, is performed. A diffusion scheme to treat viscosity at intermediate Reynolds number is formulated and analyzed in a distribution-based description. The implementation of the diffusion scheme is achieved by performing interpolation, which is a process that has been widely used to regularize particle distributions in vortex simulations, with a new set of interpolation kernels. These kernels provide an accurate and efficient way to simulate vorticity diffusion in transverse jets. An improved formulation of the vorticity flux boundary conditions is rigorously derived. / (cont.) This formulation includes separation of the wall boundary layer and feedback from the jet to the wall boundary layer, and describes detailed near-field jet structures. The results present the underlying mechanisms by which vortical structures evolve. Transformation of the jet shear layer emanating from the nozzle starts with jet streamwise lift-up of its lee side to form sections of counter-rotating vorticity aligned with the jet trajectory. Periodic rollup of the shear layer, which is similar to the Kelvin-Helmholtz instability in free shear layers, accompanies this deformation. A sudden breakdown of these coherent structures into dense vortical structures of smaller scales is observed. This breakdown to small-scale structures is due to the interaction of counter-rotating vortices and rolled-up shear layer. With a separated wall boundary layer, strong near-wall counter-rotating vortices are observed. This observation substantiates the importance of including the full interaction between the wall boundary layer and the jet shear layer in the investigation of transverse jet dynamics. / by Daehyun Wee. / Sc.D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2007. / Includes bibliographical references (leaves 28-29). / The Mars Gravity Biosatellite will house fifteen mice in a low Earth orbit satellite spinning about its longitudinal axis. The satellite's payload thermal control system will reject heat through the base of the payload module and provide air circulation vital to maintaining a habitable environment for the mice. The centripetal acceleration due to rotation creates the tendency for heated air to move by free convection toward the axis of rotation. Dominance of forced convection throughout the payload module will ensure nearly isothermal air and effective heat rejection from the payload to the bus module via fan/heatsink/thermoelectric cooler units. Circulation effectiveness is measured by the Richardson number, which expresses the ratio of the influence of free convection to the influence of forced convection in a mixed-convection flow. Experiments were executed with the current circulation system to determine the forced convection flow velocity. The free convection flow parameter was determined theoretically. Cross-flow fan/heatsink units mounted on the baseplate rim created low Reynolds number (88-985) flow throughout the enclosure. The calculated Richardson number for the worst-case 19°C difference between heated components and cooled air is between 0.78 and 2.34. / (cont.) For a realistic steady 3°C-80C temperature difference, the calculated Richardson number for the overall flow field is between 0.22 and 0.37. It was found that the flow capacity of the fan/heatsink assemblies must be increased from 1CFM to 5CFM to achieve the desired dominance of forced convection (a Richardson number of 0.1) in the worst-case on-orbit scenario. Increasing the capacity of the circulation system would allow for recovery from worst-case thermal scenarios under spaceflight conditions and allow a fraction of the cooler units to be powered during normal spaceflight conditions. The methods used here are scalable to analysis for the design of rotating human habitation vehicles. / by Jesse B. Marsh. / S.B.