Search

The interplay of bulk and boundary scattering is explored in a regime where quantum size effects modify mesoscopic transport in a degenerate Fermi liquid film of $^3$He on a rough surface. We discuss mass transport and the momentum relaxation time of the film in a torsional oscillator geometry within the framework of a quasiclassical theory that includes the experimentally determined power spectrum of the rough surface. The theory explains the anomalous temperature dependence of the relaxation rate observed experimentally. We model further studies on $^3$He confined in nanofluidic sample chambers with lithographically defined surface roughness. The improved understanding of surface roughness scattering can be extended to the analogous system of electrons in metals and suggests routes to improve the conductivity of thin metallic films., Comment: 6 pages, 1 figure

We demonstrate the fabrication of $\sim$1.08 $\mu$m deep microfluidic cavities with characteristic size as large as 7 mm $\times$ 11 mm or 11 mm diameter, using a silicon$-$glass anodic bonding technique that does not require posts to act as separators to define cavity height. Since the phase diagram of $^3$He is significantly altered under confinement, posts might act as pinning centers for phase boundaries. The previous generation of cavities relied on full wafer-bonding which is more prone to failure and requires dicing post-bonding, whereas the these cavities are made by bonding a pre-cut piece of Hoya SD-2 glass to a patterned piece of silicon in which the cavity is defined by etching. Anodic bonding was carried out at 425 $^{\circ}$C with 200 V, and we observe that pressurizing the cavity to failure ($>$ 30 bar pressure) results in glass breaking, rather than the glass-silicon bond separation. In this article, we discuss the detailed fabrication of the cavity, its edges, and details of the junction between the coin silver fill line and the silicon base of the cavity that enables a low internal-friction joint. This feature is important for mass coupling torsional oscillator experimental assays of the superfluid inertial contribution where a high quality factor ($Q$) improves frequency resolution. The surface preparation that yields well-characterized smooth surfaces to eliminate pinning sites, the use of transparent glass as a cover permitting optical access, low temperature capability and attachment of pressure-capable ports for fluid access may be features that are important in other applications., Comment: 10 figures

The influence of confinement on the topological phases of superfluid 3He is studied using the torsional pendulum method. We focus on the phase transition between the chiral A-phase and the time-reversal-invariant B-phase, motivated by the prediction of a spatiallymodulated (stripe) phase at the A-B phase boundary. We confine superfluid 3He to a single 1.08 {\mu}m thick nanofluidic cavity incorporated into a high-precision torsion pendulum, and map the phase diagram between 0.1 and 5.6 bar. We observe only small supercooling of the A-phase, in comparison to bulk or when confined in aerogel. This has a non-monotonic pressure dependence, suggesting that a new intrinsic B-phase nucleation mechanism operates under confinement, mediated by the putative stripe phase. Both the phase diagram and the relative superfluid fraction of the A and B phases, show that strong coupling is present at all pressures, with implications for the stability of the stripe phase., Comment: 6 figures, 1 table + supplemental information

We present data for energy dissipation factor (Q^{-1}) over a broad temperature range at various pressures of a torsion pendulum setup used to study 3He confined in a 98% open silica aerogel. Values for Q^{-1} above T_c are temperature independent and have a weak pressure dependence. Below T_c, a deliberate axial compression of the aerogel by 10% widens the range of metastability for a superfluid Equal Spin Pairing (ESP) state; we observe this ESP phase on cooling and the B phase on warming over an extended temperature region. While the dissipation for the B phase tends to zero as T goes to 0, Q^{-1} exhibits a peak value greater than that at T_c at intermediate temperatures. Values for Q^{-1} in the ESP phase are consistently higher than in the B phase and are proportional to \rho_s/\rho until the ESP to B phase transition is attained. We apply a viscoelastic collision-drag model, which couples the motion of the helium and the aerogel through a frictional relaxation time \tau_f. Our dissipation data is not sensitive to the damping due to the presumed small but non-zero value of \tau_f. The result is that an additional mechanism to dissipate energy not captured in the collision-drag model and related to the emergence of the superfluid order must exist. The extra dissipation below T_c is possibly associated with mutual friction between the superfluid phases and the clamped normal fluid. The pressure dependence of the measured dissipation in both superfluid phases is likely related to the pressure dependence of the gap structure of the "dirty" superfluid. The large dissipation in the ESP state is consistent with the phase being the A or the Polar with the order parameter nodes oriented in the plane of the cell and perpendicular to the aerogel anisotropy axis., Comment: 12 pages, 7 figures

Motivated by the recent prediction that uniaxially compressed aerogel can stabilize the anisotropic A phase over the isotropic B phase, we measure the pressure dependent superfluid fraction of 3He entrained in 10% axially compressed, 98% porous aerogel. We observe that a broad region of the temperature-pressure phase diagram is occupied by the metastable A phase. The reappearance of the A phase on warming from the B phase, before superfluidity is extinguished at Tc, is in contrast to its absence in uncompressed aerogel. The phase diagram is modified from that of pure 3He, with the disappearance of the polycritical point (PCP) and the appearance of a region of A phase extending below the PCP of bulk 3He, even in zero applied magnetic field. The expected alignment of the A phase texture by compression is not observed., Comment: 4 pages, 4 figures. Submitted for review to PRL

We discuss the mass transport of a degenerate Fermi liquid $^3$He film over a rough surface, and the film momentum relaxation time, in the framework of theoretical predictions. In the mesoscopic r\'egime, the anomalous temperature dependence of the relaxation time is explained in terms of the interference between elastic boundary scattering and inelastic quasiparticle-quasiparticle scattering within the film. We exploit a quasiclassical treatment of quantum size effects in the film in which the surface roughness, whose power spectrum is experimentally determined, is mapped into an effective disorder potential within a film of uniform thickness. Confirmation is provided by the introduction of elastic scattering centres within the film. We model further studies on $^3$He confined in nanofluidic sample chambers with lithographically defined surface roughness. The improved understanding of surface roughness scattering may impact on enhancing the conductivity in thin metallic films., Comment: 5 pages, 3 figures

We present results of the $Q^{-1}$ and period shift, $\Delta P$, for $^3$He confined in a 98% nominal open aerogel on a torsion pendulum. The aerogel is compressed uniaxially by 10% along a direction aligned to the torsion pendulum axis and was grown within a 400 $\mu$m tall pancake (after compression) similar to an Andronikashvili geometry. The result is a high $Q$ pendulum able to resolve $Q^{-1}$ and mass coupling of the impurity-limited $^3$He over the whole temperature range. After measuring the empty cell background, we filled the cell above the critical point and observe a temperature dependent period shift, $\Delta P$, between 100 mK and 3 mK that is 2.9$%$ of the period shift (after filling) at 100 mK. The $Q^{-1}$ due to the $^3$He decreases by an order of magnitude between 100 mK and 3 mK at a pressure of $0.14\pm0.03$ bar. We compare the observable quantities to the corresponding calculated $Q^{-1}$ and period shift for bulk $^3$He., Comment: 8 pages, 3 figures

We have developed a nuclear magnetic resonance spectrometer with a series tuned input circuit for measurements on samples at millikelvin temperatures based on an integrated two-stage superconducting quantum interference device current sensor, with an energy sensitivty e = 26 +/-1 h when operated at 1.4K. To maximise the sensitivity both the NMR pickup coil and tuning capacitor need to be cooled, and the tank circuit parameters should be chosen to equalise the contributions from circulating current noise and voltage noise in the SQUID. A noise temperature TN = 7 +/-2 mK was measured, at a frequency of 0.884 MHz, with the circuit parameters close to optimum., Comment: 3 pages, 3 figures. The following article has been submitted to Applied Physics Letters. After it is published it will be found at http://apl.aip.org/ Typos corrected, an additional reference included