Search

Nuclear magnetism in the second layer of 3He films on graphite has been studied as a function of density in the zero-field limit at temperatures down to 300 µK. Total surface coverages in this study ranged from 20 atoms/nm2 (the lowest coverage at which the effective exchange parameter turns ferromagnetic) through the two-phase region (ending at about 24 atoms/nm2) and into the single-phase imcommensurate solid region up to 31.2 atoms/nm2. The surface layers were studied using both pulsed and cw NMR measured with our SQUID NMR system in magnetic fields of 500 µT and lower. As the surface density continues to increase in the single-phase region, the exchange constant decreases consistently with a picture of steric hindrance impeding ferromagnetic exchange. Several different observational features scale with this inferred exchange: low-field magnetization, NMR frequency shift, and the temperature of onset of significant magnetization.

We report measurements of the magnetic susceptibility of liquid 3He at saturated vapor pressure in the normal state from the superfluid transition temperature up to 1 K. Our results are comparable to those of Ramm, et al.1 at temperatures above 200 mK, but are systematically higher as T→0. We conclude that the zero-temperature limiting susceptibility is approximately 5.9% larger than reported by Ramm, et al. Consequently, the discrepancy in the liquid's diamagnetism between values expected theoretically and our experimentally measured results is essentially eliminated.

We measure the magnetic susceptibility of a group of materials commonly used in the construction of low temperature cryostats. These materials are considered non-magnetic at room temperature. We use a variable temperature susceptometer to measure samples at temperatures between 2.0 K and 12.0 K and at magnetic fields between 0.25 Tesla and 4 Tesla. The samples we use include Aluminum, Beryllium Copper, fused silica, Grafoil, stainless steel, filled and unfilled epoxy, Macor, polycarbonate, aerogel, silver powder, Teflon, and Vespel. We present plots of the magnetization vs. temperature and field, and a table of the parameters to a fitting function which can be used to interpolate the data.

3 He films on graphite provide a model system for 2D magnetism. At low temperatures, the thermodynamic properties of these films are dominated by exchange interactions. However more subtle interactions such as dipolar coupling and other anisotropies may eventually play a substantial role, especially in small magnetic fields. Our recent experiments use a SQUID system for both pulsed and continuous wave NMR at very low applied fields (B

We have measured the absolute thermodynamic magnetization of 3 He in all of its liquid phases while using a new calibration technique which relies on the nuclear paramagnetism over the entire temperature range below 1 K as measured by Ramm, et al. We have directly measured the temperature dependence separately of the nuclear paramagnetism via NMR and of the total magnetism via our free-surface modulation technique. We report substantial agreement with Ramm, et al. In consequence, we find that the diamagnetism of liquid 3 He is significantly smaller than expected theoretically, a condensed matter effect not seen previously. We have also remeasured the magnitude of the magnetization discontinuity at the superfluid A-B transition and find close agreement at saturated vapor pressure with NMR-deduced values from Scholz.

We have measured the thermodynamic magnetization in the various phases of normal and superfluid3He. The A-B discontinuity differs from that inferred from nuclear-specific (NMR) experiments. We offer reasons why this apparent disagreement can be consistent with our current picture of super-fluid3He. One consequence of this measurement is an improved set of values for the phenomenological Landau-Ginzburg free energy β-parameters.

We present data for the magnitude of the thermodynamic magnetization discontinuity at the equilibrium A-B transition in superfluid3He, ΔMAB, over its entire pressure-temperature-magnetic field phase diagram. The data was taken by measuring flux changes in a SQUID pick-up coil, which was calibrated by a novel first-principles technique. We compare our results with those of NMR experiments, and find that the two types of measurements yield different values, and this discrepancy is not the result of thermometry or field errors.

We have studied the quantum corrections to the resistance of one-dimensional gold wires with periodically spaced dangling side branches. Both the weak localization and interaction corrections to the resistance of such samples are strongly reduced when the distance between branches becomes smaller than the quantum length scales. The dependence of the quantum corrections on the distance between the branches then becomes a measurement of the quantum lengths

We have studied the quantum corrections to the resistance of two-dimensional (2D) percolation networks of quasi-one-dimensional (1D) wires in which the ratio between the percolation correlation length and the quantum lengths L,LT could be controllably changed over a wide range. We have shown that the percolation networks behave like homogeneous 2D conductors with respect to the quantum corrections if L, LT. Close to the percolation threshold (L, LT) a crossover to 1D quantum corrections is observed

Transport measurements are performed under the variations of temperature and magnetic field on single crystalline InN nanowires. Conduction at low temperature reveals a three-dimensional Mott variable range hopping mechanism. With rising temperature, a semiconductor-to-metal transition is observed around 80 K. In addition, the nanowire exhibits negative magnetoresistance under both parallel and perpendicular fields due to the suppression of the electron wave function interference. A field direction asymmetry on the change of magnetoresistance is examined, attributing to the conduction channel anisotropy.

The temperature dependence of the resistivity p of thin gold films (thickness d=100-400 A) has been measured at T=30 mK-300 K. In a wide temperature range below θ D /15 (θ D is the Debye temperature) Δρ(T)/p is proportional to T 2 and does not depend on the mean free path of electrons. Experimental determinations of the dependences Δρ(T)/ρ in this temperature range are in good agreement with the correction to the impurity resistivity of a normal metal due to the quantum interference between the electron-phonon and electron-impurity interactions

Three small superconducting magnets with active and passive shields that produce confined magnetic fields in low temperature cryostats have been designed and built. The design of these magnets uses both on and off-axis field calculations which are fed back to an optimization algorithm. These design principles and construction techniques are discussed.

We have studied thin disordered metal films at low temperatures under nonequilibrium conditions. The electron temperature was raised above the lattice temperature by applying a dc bias. The Coulomb anomaly and weak localization were used as thermometers for the electrons. We confirm that at temperatures below 1 K the electron temperatures as derived from either the Coulomb anomaly or weak localization are consistent. The electron-phonon energy relaxation rate was measured to be proportional to T 3

The early studies of the superfluid phases of3He were marked by often spectacular agreements between theoretical predictions and experimental results. Nevertheless, several experiments have failed to achieve either agreement with the prevailing theory or consistency with each other — and improved experiments do not remove these discrepancies. We will discuss two recent experiments at U.S.C. which indicate that there remain fundamental questions beyond experimental uncertainties. One of these is the stability of superflow in a complex texture in the A-phase. Experimental evidence indicates that superflow may actually be stabilized by the existence of a texture. The other is the long-standing discrepancy between static (SQUID) and dynamic (NMR) measurements of the magnetization discontinuity at the A-B transition. Our recent experiments using a first principles measurement of ΔM support the continued existence of the discrepancy.

Using a sensitive differential pressure gauge, we have measured the limits to dissipationless superflow in bulk superfluid 3He-A far from stabilizing walls. We report the first quantitative measurements of the breakdown of bulk superflow in the A phase. In addition to confirming the existence of a finite critical current in bulk uniform textures, we have discovered an unexpected tortuous texture in which there is an even larger critical current. These results indicate that the current theoretical models for determining textures are incomplete.

We have measured the initial magnetic suppression of the {ital B} phase of the superfluid {sup 3}He at pressures from zero to 29 bars. This suppression, which is asymptotically quadratic in the field strength, is always significantly greater than the weak-coupling prediction, even at zero pressure. We also provide the first values of the {ital A}-phase specific-heat jump for pressures below the polycritical point. Microscopic models of {ital normal} liquid {sup 3}He do not conform to these experimental constraints. Consequently, it is possible that the conventional identification of the order parameter of superfluid {sup 3}He-{ital A} is incorrect.

Mesure de la resistance du contact thermique a travers l'interface entre PrNi 5 et 3 H e . Sur tout l'intervalle de temperatures etudiees (2-15 mK), la resistance du contact varie comme T −2 . En presence de champs magnetiques faibles (0-40 mT) la resistance du contact diminue de facon significative avec l'augmentation de l'intensite du champ. Discussion de la capacite calorifique de PrNi 5 deduite des constantes thermiques mesurees. Comparaison avec d'autres resultats

In earlier experiments on ferromagnetic 3He films, we observed a complex lineshape due in part to the dipolar field generated by polarization of the 3He nuclei. Much of the complex lineshape can be explained by the known distribution of the Grafoil platelets. However, there remained some evidence for a split NMR line at some temperatures. In our new experiments on ZYX grade exfoliated graphite where the size of individual platelets is much larger and the angular distribution is three times smaller, this splitting has become more evident over a wider range of temperatures. Now it is clear that the complex lineshape includes two peaks along with remaining orientation effects. We also find that roughly 2% of our signal comes from randomly oriented platelets. We present the details of our model for analyzing these lineshapes and the experimental results for the line splitting at several coverages in the ferromagnetic range. We discuss the possible sources of this line splitting.

We have fabricated and characterized a new type of electrometer that couples two parallel single‐electron transistors (SETs) to a radio‐frequency tank circuit for use as a differential RF‐SET. We demonstrate operation of this device in summing, differential, and single‐SET operation modes. In differential mode, the device is sensitive to uncorrelated input signals while screening out correlated ones.

What is the effect of the structural length scale on the ordering of 3He films? NMR experiments on the magnetism of second layer 3He on Grafoil in the low field limit found ferromagnetic ordering for coverages over 20 atoms/nm2. Finite temperature phase transitions are prohibited in 2D when only Heisenberg interactions are present. However ordering of a two‐dimensional magnetic film can be a result of a phase transition caused by weak anisotropy and/or dipolar interactions, or could be a less interesting manifestation of finite size effects. By replacing Grafoil with ZYX grade highly oriented graphite, we can study the magnetism of two‐dimensional films with a substantially increased structural coherence length and test the importance of finite size effects. Our new experiments find a region of coverages where the second layer 3He films become ferromagnetic at temperatures above 1 mK, with no evidence for an increased suppression of the ordering due to increasing the coherence length. We show the results ...

We investigate the pheonomenon of quasiparticle poisoning in a single Cooper‐pair box (SCB). We have designed, fabricated, and tested an SCB that demonstrates a transition between poisoned and unpoisoned Coulomb staircases, depending on the speed with which the gate charge is swept. Poisoning is shown to be suppressed at moderately high sweep rates. Coulomb staircases were measured for a variety of sweep rates, and quasiparticle tunneling rates were extracted from this data.

We have fabricated a single Cooper-pair box (SCB) in close proximity to a single electron transistor (SET) operated in the radio-frequency mode (RF-SET) with an inductor and capacitor lithographed directly on chip. The RF-SET was used to measure the charge state of the SCB revealing a 2e periodic charge quantization. We performed spectroscopy measurements to extract the charging energy (E C ) and the Josephson coupling energy (E J ). Control of the temporal evolution of the quantum charge state was achieved by applying fast dc pulses to the SCB gate. The dephasing and relaxation times were extracted from these measurements.

We provide evidence for a finite-temperature ferromagnetic transition in two dimensions as $H\ensuremath{\rightarrow}0$ in thin films of ${}^{3}\mathrm{He}$ on graphite, a model system for the study of two-dimensional magnetism. We perform pulsed and cw NMR experiments at fields of 0.03--0.48 mT on ${}^{3}\mathrm{He}$ at areal densities of $20.5--24.2\mathrm{atoms}/{\mathrm{nm}}^{2}$. At these densities, the second layer of ${}^{3}\mathrm{He}$ has a strongly ferromagnetic tendency. With decreasing temperature, we find a rapid onset of magnetization that becomes independent of the applied field at temperatures in the vicinity of 1 mK. Both the dipolar field and the NMR linewidth grow rapidly as well, which is consistent with a large (order unity) polarization of the ${}^{3}\mathrm{He}$ spins.

We provide evidence for a finite-temperature ferromagnetic transition in two dimensions as H --0 in thin films of 3He on graphite, a model system for the study of two-dimensional magnetism. We perform pulsed and cw NMR experiments at fields of 0.03-0.48 mT on 3He at areal densities of 20.5-24.2 atoms/nm(2). At these densities, the second layer of 3He has a strongly ferromagnetic tendency. With decreasing temperature, we find a rapid onset of magnetization that becomes independent of the applied field at temperatures in the vicinity of 1 mK. Both the dipolar field and the NMR linewidth grow rapidly as well, which is consistent with a large (order unity) polarization of the 3He spins.

Recent experiments at USC have shown that two-dimensional films of 3 He on graphite order ferromagnetically at finite temperatures ( T >0) for densities above 20 atoms / nm 2 . These results, obtained by measuring NMR on these films in the zero field limit, appear to contradict several studies which concluded that there is no spontaneous order by measuring magnetic susceptibility through conventional NMR in finite magnetic fields, and heat capacity in zero field. The low field limit is important in understanding these two-dimensional magnetic systems. At higher temperatures and/or fields, these films have been described in terms of the multiple spin exchange model. However, two-dimensional systems are extremely sensitive to anisotropies such as nuclear dipole interactions, even when they are several orders of magnitude less than the dominant exchange mechanism.

We use SQUID NMR to observe the magnetization of 3 He films with densities between 20 and 24 atoms / nm 2 in the zero field limit. Ferromagnetism in these nearly 2D Heisenberg exchange systems is stabilized by weak anisotropies. In the ferromagnetic phase, the NMR line becomes very broad and shifts to lower frequencies, consistent with a large dipolar field opposing a perpendicular applied field. Grafoil can be modeled by a Gaussian distribution of platelet angles centered on the normal to the Grafoil plus a randomly distributed set of platelets. Using the spin dynamics of a 2D polarized sheet with reasonable assumptions for the distributions of platelet angles, we show that the magnetization, frequency shift and line shape form a consistent picture of a highly polarized sheet.

Conflicting results between different laboratories of measurements of the phase diagram of superfluid 3 He when immersed in aerogel, even with nominally identical aerogels, suggest that details of the aerogels other than the porosity are important in determining superfluid properties. We have simulated the growth of small clusters ðNB10 5 Þ by offlattice single-particle diffusion limited aggregation (DLA). Measurements of the free-path distribution function of 3 He quasiparticles scattered by these clusters—truncated and repeated to simulate experimental conditions—show significant differences from uncorrelated scatterers at the same density, which differences may be significant for superfluid properties. r 2003 Elsevier Science B.V. All rights reserved.

We present measurements of hysteretic resistive SQUIDs (R-SQUIDs) as a function of the hysteresis parameter β L =2 πLI c / Φ 0 . Varying β L in situ from 15 to 45, we observe quasiperiodic peaks in frequency noise which decrease toward the thermal limit with increasing temperature. We have developed a model illustrating the dynamics of the two characteristic time scales: Josephson phase oscillations and catastrophic transitions in phase. From this model, an intuitive energy picture emerges wherein low thermal energy is shown to promote trapping in false minima leading to excess frequency noise. The energy model further illustrates that confusion between degenerate branch solutions results in quasiperiodic noise peaks.

We have measured the magnetoresistance of narrow (600\char21{}1000 \AA{}) and thin (150\char21{}200 \AA{}) gold films, one dimensional with respect to weak localization and electron-electron interaction effects. It is shown that electron-electron collisions with small energy transfer (Nyquist phase-breaking mechanism) govern the phase relaxation in such films over a wide temperature range. The Nyquist time ${\mathrm{\ensuremath{\tau}}}_{\mathit{N}}$ was estimated from the magnetoresistance data on the basis of the theoretical dependence \ensuremath{\Delta}${\mathit{R}}_{\mathrm{loc}}$(H) that is applicable to the case when Nyquist phase relaxation dominates other phase-breaking proceses. The temperature dependence ${\mathrm{\ensuremath{\tau}}}_{\mathit{N}}$(T)\ensuremath{\propto}${\mathit{T}}^{\mathrm{\ensuremath{-}}2/3}$ obtained in this way is in good agreement with the theoretical prediction for one-dimensional conductors.

We have measured the absolute magnetization of liquid 3He at saturated vapor pressure over the temperature range 0.3–500 mK. The technique compares the magnetization of the liquid to that of vacuum by passing the liquid–vacuum interface through a SQUID coil. We find that, contrary to expectations, the electronic diamagnetism of the normal liquid is approximately 6% less than the calculated atomic value. This appears to be the first measurement of the diamagnetic susceptibility of any noble gas in a condensed state.

3He films on graphite at densities above 20 atoms / nm 2 have a strong ferromagnetic tendency. Models of multiple spin exchange are believed to describe the dominant interaction. However, other much smaller interactions can significantly alter the low-temperature behavior of these systems, especially in very low magnetic fields. Our recent experiments have used a SQUID system for both pulsed and continuous wave NMR at very low applied fields (B mT ) . Our results show that the polarization of this system grows much more rapidly than is expected from the Heisenberg exchange models, even approaching saturation. Possible explanations are the influence of finite size effects and dipolar interactions.

The use of a SQUID coupled to a sample by a superconducting circuit allows NMR measurements over wide bandwidths. For our recent NMR experiments on 3He monolayers we developed a sample cell that allows for indirect cooling of the grafoil without excessive eddy currents. The details of the cell design are discussed. In addition, we have found that the 13C signal from the graphite is observable in this system. The FID time constant for this signal is comparable to the field homogeneity and the amplitude follows the Curie law over our temperature range ( 15 mK >T>0.2 mK ). Since the NMR signal in 13C is observed simultaneously with the 3He signal, this signal can act as a thermometer for the graphite substrate.

The boundary layer of 3He on graphite has shown properties of a two-dimensional ferromagnet. Both, in the cases of 3He filling pores of Grafoil and with a few monolayers, magnetization far in excess of the Curie value are observed at millidegree temperatures. Our experiments on the low-field NMR spectra of the 3He boundary layers in filled pores show collective modes for T < 1 mK. One of these modes extrapolates to nonzero frequency and amplitude as the field is dropped to zero. The Heisenberg symmetry is broken by the weak dipolar energy. This dipolar energy and finite size effects allow for order in zero field and finite temperature.

Krotscheck to Walraven : Normally one sees interesting effects when the energy of the probe agrees with the energy of one of the collective modes. In helium films, one would expect such a level crossing at about 1 K energy of the H-atom with the ripplon, and at larger or equal to 10 K with the phonon/roton. Is it possible to make experiments in that regime of energies of the H-atoms?

At low temperatures, the quantum corrections to the resistance due to weak localization and electron-electron interaction are affected by the shape and topology of samples. We observed these effects in the resistance of 2D percolation networks made from 1D wires and in a series of long 1D wires with regularly spaced side branches. Branches outside the classical current path strongly reduce the quantum corrections to the resistance and these reductions become a measure of the quantum lengths.

Combinations of the Landau-Ginzburg phenomenological β-parameters characterizing superfluid3He have been deduced from several thermodynamic measurements. These β-parameters impose constraints on microscopic models of normal3He, and select the structure of the microscopic order parameter in the superfluid.

The absolute thermodynamic magnetization of liquid3He has been measured from 0.5 mK–100 mK at saturated vapor pressure. The measurement uses a SQUID magnetometer and a simple displacement technique to reference the total magnetization of the liquid to an absolute vacuum level. Because the nuclear paramagnetic magnetization equilibrates on a time scale long compared to that of the core diamagnetic magnetization, we are able to separate these two signals throughout much of the normal Fermi liquid regime. The diamagnetic component dominantes the paramagnetic over the entire temperature range.

We report results from a new SQUID-based measurement of the total magnetization change of superfluid 3 He at the A-B phase transition, ΔM AB total . We used a simple first-principles calibration technique. At low fields ∼0.03 T, our data agree with past measurements, verifying the long-standing discrepancy between ΔM AB total and ΔM AB nuclear , the magnetization change which is measured by NMR. Use of a self-shielded solenoid allowed the first measurement of ΔM AB total over the full range of magnetic field. We found the magnetization discrepancy to persist through the whole phase diagram.

We have measured the A-B phase transition temperature of superfluid3He at pressures from zero to 29 bars, and in all magnetic fields up to the highest field limit of the B phase (∼0.59 Tesla). We have also devised a simple function which accurately describes the entire pressure-temperature-magnetic field phase diagram of the A-B transition. This function intrinsically satisfies all correct asymptotic behaviors in the low- and high-field limits. Using thermodynamic identities we can measure properties which previously could not be directly measured.

Quantum corrections (QC) to the resistivity due to weak localization (WL) and electron-electron interaction (EEI) have been measured for percolated metal networks with percolation coherence length z comparable to the quantum length scales-the phase-breaking length L and the thermal diffusion length LT. Far from the percolation metal-insulator transition (MIT) the `macroscopic' disorder is averaged over the quantum lengths (L,LTz) and the QC are of the universal two-dimensional (2D) type with R=R(z), where R(z) is the resistance per square at scales larger than z. A crossover 2D1D in the dimensionality of the QC and the Nyquist phase-braking mechanism is observed near the MIT, where zL,LT

We have measured the low temperature magnetoresistance and Hall Voltage on an inverted high electron mobility transistor (HEMT) structure with ultra low carrier concentration in the dark . We have observed Shubnikov-de Haas oscillations and quantized Hall voltages at very low magnetic fields.

We are measuring the boundary between the superfluid A and B phases of 3He as a function of magnetic field and pressure. The results yield new information on Landau-Ginzburg parameters, microscopic quasiparticle scattering amplitudes, superfluid specific heats, and potentially other normal liquid properties through comparison with models of the superfluid.

We have searched for the pressure drops which are a necessary consequence of the dissipation attendant textural motion in 3HeA when uniform bulk superflow exceeds a critical value. We find no reliable pressure signature at a level significantly smaller than expected theoretically. We have modified our sensitive differential pressure gauge by replacing the plastic diaphragm with an extremely thin aluminum film, increasing the resolution by two orders of magnitude. Further investigations are underway.

Measurements of the magnetic susceptibility ($\ensuremath{\chi}$) of quinolinium ${(\mathrm{tetracyanoquinodimethane})}_{2}$ are reported over the temperature range $0.45 \mathrm{mK}lTl40 \mathrm{mK}$ with use of low-field ESR. Down to 7 mK, it is found that $\ensuremath{\chi}\ensuremath{\propto}{T}^{\ensuremath{-}\ensuremath{\alpha}}$ with $\ensuremath{\alpha}=0.81$ characteristic of a random-exchange Heisenberg antiferromagnetic chain (REHAC). At lower temperatures, the susceptibility peaks, the ESR linewidth greatly increases, and an internal field rapidly builds up. These changes may be due to either hyperfine or dipolar interactions, signalling a crossover from one-dimensional REHAC to a three-dimensional behavior-possibly a dipolar spin-glass.

Magnetic textures have been created and studied in the A phase of superfluid 3He at several pressures, in magnetic fields from 15 to 20 mT, by intense rf magnetic pulses which invert the magnetization of the fluid. To create this new state we find that the initial large pulse must tip the magnetization by at least 140‡. In this state the fluid evolves from a condition where the NMR frequency shift Ω-γH 0is smaller than that in the unperturbed superfluid by a factor of 0.54, over a time scale of tens of seconds to a condition where the frequency shift is 0.70 times the unperturbed value. We have studied the fluid's linear response to small-angle sampling pulses as well as the nonlinear response to large-angle pulses. By using two independent NMR coils we have seen this new state move through the fluid; we find that a lower limit on its propagation velocity is around one-half the spin wave velocity. All of our results are generally consistent with the model of solitons in the order parameter proposed by Maki and co-workers.

A propagating magnetic texture has been observed in superfluid /sup 3/He-A confined to 17-..mu..m slabs using pulsed NMR. The results support a model of an expanding lattice of solitons in the order parameter. The propagation velocity approaches the spin-wave velocity as T..-->..T/sub c/, and goes to zero for T