No abstract

Longitudinal sound attenuation measurements in superfluid 3 He in 98% aerogel were conducted at pressures between 14 and 33 bar and in magnetic fields up to 0.444 T. The temperature dependence of the ultrasound attenuation in the A-like phase was determined for the entire superfluid region by exploiting the field-induced metastable A-like phase at the highest field. In lower fields, the A-B transition in aerogel was identified by a smooth jump in attenuation on both cooling and warming. Based on the transitions observed on warming, a phase diagram as a function of pressure ͑P͒, temperature ͑T͒, and magnetic field ͑B͒ is constructed. We find that the A-B phase boundary in aerogel recedes in a drastically different manner than in bulk in response to an increasing magnetic field. The implications of the observed phase diagram are discussed.

We have performed longitudinal ultrasound (9.5 MHz) attenuation measurements in the B phase of superfluid 3He in 98% porosity aerogel down to the zero temperature limit for a wide range of pressures at zero magnetic field. The absolute attenuation was determined by direct transmission of sound pulses. Compared to the bulk fluid, our results revealed a drastically different behavior in attenuation, which is consistent with theoretical accounts with gapless excitations and a collision drag effect.

We present measurements of an amplifier based on a dc superconducting quantum interference device (SQUID) with submicron Al-AlOx-Al Josephson junctions. The small junction size reduces their self-capacitance and allows for the use of relatively large resistive shunts while maintaining nonhysteretic operation. This leads to an enhancement of the SQUID transfer function compared to SQUIDs with micron-scale junctions. The device layout is modified from that of a conventional SQUID to allow for coupling signals into the amplifier with a substantial mutual inductance for a relatively short microstrip coil. Measurements at 310 mK exhibit gain of 32 dB at 1.55 GHz.In recent years there have been many advances with amplifiers based on dc superconducting quantum interference devices (SQUIDs) with a resonant stripline input circuit [1]. In these devices the signal is coupled to the SQUID through the λ/2 microstrip resonance formed by a superconducting spiral input coil above the dielectric layer on top of the superconducting washer that forms the SQUID loop. Such amplifiers have exhibited gains in excess of 20 dB in the radiofrequency range [2] and noise temperatures at 500 MHz within a factor of two of the quantum limit [3]. In addition, microstrip SQUID amplifiers have been demonstrated at frequencies up to 7.4 GHz [4]. This suggests the possibility of using these devices for measuring the weak signals involved in various quantum information processing schemes with superconducting circuits [5], including dispersive readout with circuit quantum electrodynamics [6] and schemes involving pulsed interactions between qubits and oscillators [7]. However, the shorter coils required to increase the operating frequency lead to decreased gain: 12 dB at 2.2 GHz and 6 dB at 7.4 GHz [4]. An alternative configuration with a small-area SQUID coupled in a lumped-element configuration to a quarter-wave resonator was shown to operate as an amplifier in the GHz range [8].The gain G of a microstrip SQUID amplifier is proportional to M 2 i V 2 Φ , where M i is the mutual inductance between the input coil and the SQUID loop and V Φ ≡ ∂V /∂Φ is the maximum voltage modulation of the SQUID. Pushing the operating frequency f 0 higher requires shorter coils, which necessarily reduces M i , although this reduction can be mitigated somewhat by modifying the SQUID loop and coil layout from the conventional washer design. Nonetheless, G will decrease as f 0 is increased unless one can simultaneously compensate by increasing V Φ . The peak-to-peak voltage modulation of a SQUID is limited by the I 0 R product of each junction, where I 0 and R are the junction critical current and shunt resistance, respectively. Nonhysteretic device operation requires a junction damping pa- * bplourde@phy.syr.edu rameter β C ≡ (2πI 0 R/Φ 0 )RC ≤ 1, where Φ 0 ≡ h/2e, thus placing an upper limit on R, where C is the junction self-capacitance. For Josephson junctions fabricated with conventional photolithography with an area of a few µm 2 , C is typically a few hundred fF....