We describe a novel single-coil mutual inductance technique for measuring the magnetic penetration depth of superconductors at 2-4 MHz as a function of temperature in the 4-100 K range. We combine a single-coil configuration with a high-stability marginal oscillator; this enables us to measure the absolute value of on both bulk samples and thin films with very high resolution (␦ϭ10 pm͒ and a precision of 30 nm. As example of application, we report measurements on NbTi bulk samples and Nb films. This contactless technique is suited for probing the superconducting properties of samples over large surfaces.

Narrow YBa 2 Cu 3 O 7 films, excited by pulses of supercritical current, had their nanosecond electric response monitored in zero applied magnetic field. Delayed voltage steps plus constant differential resistance, characteristic of phase-slip centers (PSC), are observed at all temperatures. The duration of the initial zero voltage state is well fit by Ginzburg-Landau based theories, with a gap relaxation time controlled by phonon escape. At higher levels of excitation, PSC's give birth to slowly spreading normal hot spots. [S0031-9007(98)06897-5] PACS numbers: 74.76. Bz, 74.25.Kc, 74.40. + k, 74.50. + r Superconductivity is perturbed, not suppressed, by a critical current. Thanks to vortex motion, the growing phase differences of the order parameter wave function relax by multiples of 2p [1], a process which allows the material to preserve long range order while sustaining a voltage. In filamentary structures, similar quantum phase jumps occur in short, fixed, zones named phaseslip centers [2] (PSC). So, dissipation arises in a way fundamentally different from a transition into the normal state. A second singular feature is the long persistence of the zero voltage state after the application of the supercritical current, much beyond the natural (picosecond) time h͞D (Planck constant divided by the energy gap). Pals and Wolter [3] discovered this delay, or PSC nucleation time, t d on aluminum film strips. Interpreting it as the time required to achieve complete collapse of the order parameter, and using a timedependent Ginzburg-Landau (TDGL) equation based on the nonequilibrium energy-mode gap relaxation time [4], they could account fairly for their experimental data.It would be difficult to make any predictions for cuprate superconductors, due to their utterly short coherence length. Nevertheless, the dc current-voltage ͑I-V ͒ curves of high-T c narrow bridges indeed display the steps and the expected response to microwave radiation [5] related to PSC's. This Letter reports on the first observation of the delay t d in epitaxial YBa 2 Cu 3 O 7 thin films, and its interpretation through TDGL. At the same time, we tackle directly the troublesome problem of heating in a dissipative center.Our experiments were performed in zero applied magnetic field on c-axis textured films with typical misorientation 0.5 degree of arc. Sample LL-109N is a 75 nm thick film evaporated [6] on ͓100͔ MgO with cationic composition YBa 1.88 Cu 3.04 . Our bridge has resistivity r͑300 K͒ 350 mV cm, a ratio RRR r͑300 K͒͞r͑100 K͒ 2.8, and a critical temperature T c 89 K. Samples TO-34 and TO-S35, obtained by dc hollow cathode sputtering [6], had similar electrical characteristics. Finally, sample LZCYB77 is a 35 nm thick film deposited by laser ablation [6] on ͓100͔ Si covered with a buffer layer (70 nm YSZ; 10 nm CeO 2 ). Our bridge had r͑300 K͒ 1.9 mV cm, and RRR 2.6.According to the standard PSC model [7], each phaseslip event produces a burst of quasiparticles, whose diffusion eventually determines the length of the dissipative zone. T...

Tungsten films, 0.05–1 μm thick, were deposited on silicon wafers by rf magnetron sputtering. Induced film stresses were investigated as a function of substrate temperature and background pressure in connection with microstructural observations and limited compositional analysis. Homologous substrate temperatures Ts/Tm, where Tm is the melting point of tungsten (3683 K), ranged from 0.08 to 0.15. Argon pressures investigated ranged from 1 to 7 Pa. This corresponds to the low-temperature part of Thornton’s microstructure model, within zones I and T. Two regions were distinguished for increasing values of the homologous temperature and/or decreasing values of argon pressure: (i) The first region, observed for low substrate temperature and high argon pressure, was a voided zone I microstructure with small grain size (200–400 Å). Film stresses increased from zero towards tensile values as voids disappeared. (ii) The second region, observed for higher values of Ts/Tm and/or low pAr, was a denser zone T structure, with a bimodal grain distribution (<3000 Å). After reaching a maximum tensile value σ=2×103 MPa, corresponding to the transition from zone I to zone T microstructure, the stress decreased abruptly to a constant value σ=−2.7×103 MPa. All stress values were found to be quite independent of thickness in the range from 0.05 to 1 μm. Even though β metastable phase is usually observed for zone I deposits, a quasipure α film was obtained at low temperature and high pressure (within zone I) under especially clean conditions. Independent of phase composition (α or β phase) the voided zone I was observed to be highly reactive in air. The shift with time toward compressive stress for the zone I structure was attributed to oxygen absorption in the pores. For 3-month-old films, a high oxygen content of ∼20% was measured by nuclear reaction analysis. In the denser zone T films without voids, no change in film stresses was observed over time, and oxygen content was negligible (<1%) after 3 months.

We report the detection of single electrons using a Nb 0.7 Ti 0.3 N superconducting wire deposited on an oxidized silicon substrate. While it is known that this device is sensitive to single photons, we show that it also detects single electrons with kilo-electron-volt energy emitted from the cathode of a scanning electron microscope with an efficiency approaching unity. The electron and photon detection efficiency map of the same device are in good agreement. We also observe detection events outside the active area of the device, which we attribute to sensitivity to backscattered electrons. © 2010 American Institute of Physics. ͓doi:10.1063/1.3506692͔The versatility of superconducting nanowires as single particle detectors relies on their sensitivity to the minute amount of energy required to locally induce a resistive transition. From this point of view, the latest achievements involving the detection of organic molecules 1 and photons in the infrared range 2 all derive from early experiments with ␣-particles in the million electron volt range. 3 In order to go beyond the optical resolution limit, the scanning electron microscope ͑SEM͒ working at low temperature, 4 proved useful. This technique enabled the visualization of the real size of the hot spot caused by a detection process. 5 However, the best achieved spatial resolution is limited by thermal diffusion to about 1 m and single electron detection has not been demonstrated. In this paper, we show single electron detection using a superconducting nanowire. Our superconducting single electron detector ͑SSED͒ offers a high spatial and timing resolution and we compare the electron detection efficiency map with a photon detection efficiency map, measured on the same device.The fabrication process of our superconducting nanowire has been described before. 6 It consists of a 100 nm wide, 500 m long, and 6 nm thick wire of Nb 0.7 Ti 0.3 N. The wire is folded into an 10ϫ 10 m 2 area, with a separation of 100 nm between adjacent detecting branches. One end of the wire is grounded whereas the other end is connected to a cryogenic coaxial cable used to inject a current through the structure. We measure a dc critical current I c =10 A at 4.2 K and I c = 5.2 A at 8 K. Our experimental setup consists of a cryogenic SEM. 7 The detector is mounted on a cold translation stage at T = 8 K under the electronic beam of an SEM. The SEM current I b is controlled and can be measured with a picoammeter ͑measurement uncertainty 10%͒. The energy of the incident electrons E e can be varied between 5 and 30 keV.To block low frequency 1 / f noise we use a dc-block, through which we can only bias the wire with pulses of current amplitude I and duration t d = 800 ns ͑see inset of Fig. 1͒. Each pulse is reflected by the circuit. When the current is on, the detection of an electron triggers a short pulse at the output of the system ͑duration of a few nanoseconds, see inset of Fig. 1͒. The change in baseline is caused by the limited bandwidth ͑0.1 to 1000 MHz͒ of the amplification of the o...