The results of third sound and quartz crystal microbalance measurements on thin superfluid 4He films (4–16 atomic layers) exploring the effect of surface roughness, disorder and localization are reported. The effect of surface roughness on 4He adsorption and third sound speed is studied in detail for several well parameterized CaF2 surfaces. The third sound velocity and the 4He adsorption on CaF2 both show a hysteretic behavior with the adding and removing of 4He. Adsorption to the CaF2 can be as much as ten times larger than on similar smooth surfaces and the third sound travelling on this thicker 4He film propagates up to three times slower. Substrates are fabricated that contain two-dimensional random and ordered collections of disk shaped CaF2 regions on an otherwise smooth background. The difference in the third sound speed and 4He adsorption on the two regions causes the CaF2 regions to act as third sound scatterers. The frequency dependence of the third sound propagation through the arrays is reported: pass bands and band gaps are observed on the ordered array and classical wave localization effects are observed on the disordered array.

In this thesis classical disordered spin systems, in particular, the random field Ising model (RFIM), are studied using numerical methods. The previous experimental and theoretical research on the RFIM is reviewed. Critical exponents of the RFIM are extracted at zero temperature. The heat capacity exponent α is found to be near zero. The ground states within a certain region with varying external field and disorder are portrayed, and it is speculated that if related quantities are scaled properly, ground state portraits at different system sizes are statistically similar. The performance of the Wang-Landau (WL) algorithm, which is used in our positive temperature study of the RFIM, is measured by the mean first-passage time. It is proved that the mean first, passage time of the mean field Ising model scales as N 2 log N, where N is the number of spins. The logarithmic correction term to the unbiased Markovian random walk is due to the boundaries. When applied to the Ising model with nearest neighbor interactions, the WL algorithm suffers from power law slowing down. Carefully designed cluster updating, however, can accelerate the WL algorithm and eliminate the slowing down. The WL algorithm is used in our finite temperature study of the RFIM. Physical quantities, such as the specific heat and the susceptibility, are obtained by the WL algorithm over a broad range of temperature. It is discovered that for some realizations with system size larger than 8 3, sharp peaks are present in these physical quantities. These sharp peaks result from flipping a large domain, and furthermore, the critical domains are strongly correlated to the domains found at the zero temperature transition. This observation suggests a strong version of the zero temperature fixed point picture. The MKRG approximation is employed to test the correlation between the critical states and the ground states in much larger systems. It is found that the fraction of correlated realizations drops as system size grows, while there are still significant number of correlated realizations even for systems as large as 10243. Whether substantial correlations will remain in the thermodynamic limit is unclear from our present data.

Poly(lactic acid) is an environmentally-benign biodegradable and sustainable thermoplastic material, which has found broad applications as food packaging films and as non-woven fibers. The crystallization and deformation mechanisms of the polymer are largely determined by the distribution of conformation and configuration. Knowledge of these mechanisms is needed to understand the mechanical and thermal properties on which processing conditions mainly depend. In conjunction with laser light scattering, Raman spectroscopy and normal coordinate analysis are used in this thesis to elucidate these properties. Vibrational spectroscopic theory, Flory's rotational isomeric state (RIS) theory, Gaussian chain statistics and statistical mechanics are used to relate experimental data to molecular chain structure. A refined RIS model is proposed, chain rigidity recalculated and chain statistics discussed. A Raman spectroscopic characterization method for crystalline and amorphous phase orientation has been developed. A shrinkage model is also proposed to interpret the dimensional stability for fibers and uni- or biaxially stretched films. A study of stereocomplexation formed by poly(l-lactic acid) and poly(d-lactic acid) is also presented.

A new class of Monte Carlo methods is introduced. The approach is devised to be especially useful for studying critical phenomena and phase transitions. The method can sample the critical point without a priori knowledge of the critical temperature. The method is first verified for Ising and Potts models. It is shown that the method also efficiently simulates the coexistence region near first-order transitions. The method is applied to dilute Ising models for studying disorder effects on critical phenomena. The dynamic properties of the method are investigated numerically. Finally a parallel version of the method is described and high-precision numerical results for the critical point of three-dimensional Ising models are presented.

We have studied hydrodynamic dispersion in plastic bead packs using the pulsed field gradient (PFG) NMR technique. The bead diameter was varied from 15 to 138 $\mu$m and the Peclet number Pe varied from 0 to 10$\sp3$ (the Peclet number is a dimensionless measure of the flow velocity). We studied the time dependence of both the longitudinal dispersion coefficient $D\sb{\vert\vert}$ and the transverse dispersion coefficient $D\sb{\perp}.$ We observed transitions from decreasing with time at low Pe to increasing with time at high Pe for both $D\sb{\vert\vert}$ and $D\sb{\perp}.$ We used our data to find the transition time $t\sb0$ the time required for dispersion coefficient to reach its long time value. For both $D\sb{\vert\vert}$ and $D\sb{\perp}$, we found a power-law dependence of $t\sb0$ on Pe, as has been predicted by Koch and Brady. The Pe dependence of $t\sb0$ provides information on the operative dispersion mechanisms. Our results show that both convection dispersion and boundary layer dispersion contribute to longitudinal dispersion in our experiments. However, the Pe dependence of $t\sb0$ for transverse dispersion does not agree with the theoretical prediction of Koch and Brady. We measured $D\sb{\vert\vert}$ and $D\sb{\perp}$ as a function of Pe. Our experimental results are consistent with previous results measured using conventional methods. We found that the results for longitudinal dispersion agree with Saffman's capillary tube model in our observation range. The results for transverse dispersion agree with Koch and Brady's fixed bed model to some extent, but at low Pe, the disagreement is significant. We obtained the wave-number and frequency dependent nonlocal dispersion coefficient ${\buildrel{\approx}\over{D}}\sb{\vert \vert,\perp}(q,\omega)$ from our PFG NMR data. In the local (long time and distance) limit, our results agree with previous results obtained with conventional methods and for no flow they agree with a simple model of restricted diffusion. Our results for nonlocal dispersion with flow are in reasonable agreement with Koch and Brady's calculation based on a dilute-sphere approximation to the medium.

We examine several models in statistical physics from the perspective of parallel computational complexity theory. In each case, we describe a parallel method of simulation that is faster than current sequential methods. We find that parallel complexity results are in accord with intuitive notions of physical complexity for the models studied. First, we investigate the parallel complexity of sampling Lorentz lattice gas (LLG) trajectories. We show that the single-particle LLG can be simulated in highly parallel fashion, in contrast to multi-particle lattice gases which most likely cannot. In the case of diffusion-limited aggregation (DLA), we show that a polynomial speedup is feasible even though a highly parallel algorithm probably is not. In particular, we present a polynomial-processor algorithm for generating DLA clusters that runs in a time sub-linear in the cluster mass. We relate the dynamic exponent of our parallel DLA algorithm to static scaling exponents of DLA and give numerical estimates. We investigate the parallel complexity of the invaded cluster (IC) algorithm and find that a single sweep can be carried out in highly parallel fashion but that a polynomial number of sweeps most likely cannot be compressed into a polylogarithmic number of parallel steps. We argue that quantities measured for a sub-system of size l, using the IC algorithm, should exhibit a crossover to Swendsen-Wang behavior for l sufficiently smaller than the system size L, and we propose a scaling form to describe this phenomenon. By studying sub-systems, we observe critical slowing for the $2d$ Ising and 3-state Potts models. We define the dynamic exponent of the IC algorithm according to $\tau\sb{\varepsilon,\max} \sim L\sp{z{\rm IC}}$, where $\tau\sb{\varepsilon,\max}$ is the maximum value of the energy autocorrelation time attained over all sub-system sizes for a given L. We give numerical estimates for $z\sp{\rm IC}$ for the $2d$ Ising and 3-state Potts models which result in improved upper bounds on the parallel complexity of sampling the critical points of these systems.

Since the discovery and purification of C$\sb{60}$ there has been intense research interest in fullerenes. We are exploring the adsorption behavior of $\sp4$He on fullerenes and the possibility that several of the newly discovered carbon materials provide a host environment for $\sp4$He in which the $\sp4$He properties are substantially different from the three-dimensional case. The deeper interest in this question lies in the possibility of realizing a quantum fluid in such an environment. We are examining the possibilities of $\sp4$He intercalation into C$\sb{60}$ and into bundles of Single Wall Carbon Nanotubes (SWCNs). The former provides a cubic lattice with potential intercalation into the interstitial sites. The results presented here argue that deep penetration into the C$\sb{60}$ matrix does not occur under the conditions of this study. We do however find evidence for some excess adsorption that is independent of the C$\sb{60}$ film thickness. SWCNs might allow $\sp4$He intercalation into the insides of the tubes (if the endcaps are removed) or in between the tubes (if they are arranged parallel to each other), thus allowing for quasi one-dimensional systems of $\sp4$He. The results presented here show that $\sp4$He atoms are strongly bound to SWCNs with substantial $\sp4$He presence in the vicinity of the SWCNs for T $\sim$ 15K. A model of a 1-dimensional ideal gas with a binding energy of $\sim$330K is consistent with the low temperature data presented here. Our experiments on $\sp4$He with C$\sb{60}$ films utilize Quartz Crystal Microbalances (QCMs) that were coated with C$\sb{60}.$ We measured the $\sp4$He mass adsorption as $\sp4$He was added under isothermal conditions or under conditions where the $\sp4$He charge in the sample cell was held constant as the temperature was varied. The experiments of $\sp4$He in the vicinity of SWCNs use bundles of SWCNs that were stored in vacuum since purification. We measured the $\sp4$He desorption as the temperature was linearly increased from T = 14K to T = 320K.

This work extends the study of helium films with the description of two experiments performed on different substrates. The primary experiment involves 4He and 3He-4He mixture films on a solid H2 substrate. The secondary experiment is concerned with the behavior of mixture films on Nuclepore. The substrate for the helium-on-hydrogen experiment was chosen for the weak binding potential for helium on hydrogen. Unlike most substrates, H 2 will support a superfluid 4He film less dense than a single bulk-density atomic layer. In recent years, interesting effects have been observed for helium-on-hydrogen systems. The H2 experiment is presented with the results of quartz crystal microbalance measurements and attempted third sound measurements. The quartz crystal microbalance technique measures mass coupling and reveals several interesting features in monolayer and submonolayer helium films on H2. Submonolayer superfluidity and possible phase separation of two-dimensional submonolayer 4He films on solid H2 is reported. 3He–4He mixture films on the H2 substrate reveal novel dependence of the Kosterlitz-Thouless transition temperature on 3He coverage and an unexpected feature at lower temperature. Third sound resonance and NMR measurements are reported for the Nuclepore experiment. The Nuclepore experiment is the continuation of a long-term study of multilayer 3He-4He mixture films. NMR relaxation times, magnetization, and spin diffusion measurements are discussed over a range of temperatures and mixture film coverages. Third sound measurements taken at the same mixture film coverages reveal interesting structure as a function of temperature. Third sound frequency data at temperatures near 150 mK and at 3He coverages greater than one atomic layer exhibit a local maximum that responds to changes in both 3He and 4He coverage.

Normal-Superconductor-Superconductor (NSS) single-electron tunneling (SET) transistors were studied. A hysteresis in the current versus gate voltage curve was found in a SET transistor and SET trap hybrid system. This hybrid system can be used as a new type of memory cell in electronics. The inferred junction capacitance, as extracted from the nonlinear current-voltage characteristics, was measured as a function of magnetic field and temperature. Noise issue in SET transistor was studied as well.

Investigation of helium on rubidium, sodium and potassium surfaces followed the cesium studies. My own involvement in wetting experiments was triggered by the results of an experiment conducted by Wyatt and Klier. In their experiment the helium-rubidium system seemed to exhibit nonwet behavior below 300 mK. The very thin film of adsorbed helium on nonwet rubidium seemed to behave like a submonolayer superfluid. From it was born the idea that we could explore superfluidity in the thin film of helium atop rubidium by conducting an experiment which attempts to propagate a tidal-like wave (third sound) along the film. The basic idea behind the experiment is that third sound can only propagate in films which are superfluid. Therefore, if the thin film of helium on rubidium is superfluid, we can expect to propagate third sound through it. We designed an experiment to examine third sound that propagates across a rubidium substrate. We were able to use the third sound reflection off of the rubidium boundary as a diagnostic to determine when the rubidium was wet (or prewet) by helium and when it was not wet by helium. We produced a rubidium substrate that could be caused to make a transition from being nonwet by helium to being wet by helium. Third sound propagated across the rubidium. We therefore have obtained strong evidence to support the idea that helium in the thin film state on rubidium can behave like a superfluid. We also observed strong evidence that, very close to saturation, the rubidium was always wet by helium to temperatures as low as 150 mK. On a second rubidium surface, the rubidium was always in the wet state, evidence that it is extremely difficult to fabricate rubidium surfaces of reproducible quality. In this document, we present the results of our third sound experiments on rubidium and explain how we have drawn conclusions about the wetting behavior of helium on rubidium. In addition, a chapter is devoted to some of the low temperature third sound data on glass. (Abstract shortened by UMI.)