|
|
Skocpol', W. J., Beasley, M. R., & Tinkham M. (1974). Self-heating hotspots in superconducting thin film microbridges. J. Appl. Phys., 45, 4054–4066.
Abstract: Heating effects in both long and short superconducting thin-<ef><ac><81>lm rnicrobridges are described and analyzed. Except near T(c), at low voltages where superconducting quantum processes occur, all of our experimental dc I-V characteristics can be satisfactorily understood on the basis of a simple model of a localized normal hotspot maintained by Joule heating. We consider approximations appropriate to the cases of long bridges, short bridges, and bridges coupled to microwave radiation. The analysis leads to analytic expressions for the I-V characteristics which agree well with the experimental data. We show that the formation of such a hotspot is the dominant cause of the hysteresis observed in the I-V characteristics at low temperatures. We also show that the growth of such a hotspot imposes a high-voltage limit on the ac Josephson effect in these devices, and we compare the importance of such heating effects at high voltages in various types of superconducting weak links.
|
|
|
|
Nebosis, R. S., Steinke, R., Lang, P. T., Schatz, W., Heusinger, M. A., Renk, K. F., et al. (1992). Picosecond YBa2Cu3O7−δdetector for far‐infrared radiation. J. Appl. Phys., 72(11), 5496–5499.
Abstract: We report on a picosecond YBa2Cu3O7−δ detector for far‐infrared radiation. The detector, consisting of a current carrying structure cooled to liquid‐nitrogen temperature, was studied by use of ultrashort laser pulses from an optically pumped far‐infrared laser in the frequency range from 25 to 215 cm−1. We found that the sensitivity (1 mV/W) was almost constant in this frequency range. We estimated a noise equivalent power of less than 5×10−7 W Hz−1/2. Taking into account the results of a mixing experiment (in the frequency range from 4 to 30 cm−1) we suggest that the response time of the detector was few picoseconds.
|
|
|
|
Gousev, Y. P., Gol'tsman, G. N., Semenov, A. D., Gershenzon, E. M., Nebosis, R. S., Heusinger, M. A., et al. (1994). Broadband ultrafast superconducting NbN detector for electromagnetic radiation. J. Appl. Phys., 75(7), 3695–3697.
Abstract: An ultrafast detector that is sensitive to radiation in a broad spectral range from submillimeter waves to visible light is reported. It consists of a structured NbN thin film cooled to a temperature below Tc (∼11 K). Using 20 ps pulses of a GaAs laser, we observed signal pulses with both rise and decay time of about 50 ps. From the analysis of a mixing experiment with submillimeter radiation we estimate an intrinsic response time of the detector of ∼12 ps. The sensitivity was found to be similar for the near‐infrared and submillimeter radiation. Broadband sensitivity and short response time are attributed to a quasiparticle heating effect.
|
|
|
|
Danerud, M., Winkler, D., Lindgren, M., Zorin, M., Trifonov, V., Karasik, B. S., et al. (1994). Nonequilibrium and bolometric photoresponse in patterned YBa2Cu3O7−δ thin films. J. Appl. Phys., 76(3), 1902–1909.
Abstract: Epitaxial laser deposited YBa2Cu3O7−δ films of ∼50 nm thickness were patterned into detectors consisting of ten parallel 1 μm wide strips in order to study nonequilibrium and bolometric effects. Typically, the patterned samples had critical temperatures around 86 K, transition widths around 2 K and critical current densities above 1×106A/cm2 at 77 K. Pulsed laser measurements at 0.8 μm wavelength (17 ps full width at half maximum) showed a ∼30 ps response, attributed to electron heating, followed by a slower bolometric decay. Amplitude modulation in the band fmod=100 kHz–10 GHz of a laser with wavelength λ=0.8 μm showed two different thermal relaxations in the photoresponse. Phonon escape from the film (∼3 ns) is the limiting process, followed by heat diffusion in the substrate. Similar relaxations were also seen for λ=10.6 μm. The photoresponse measurements were made with the film in the resistive state and extended into the normal state. These states were created by supercritical bias currents. Measurements between 75 and 95 K (i.e., from below to above Tc) showed that the photoresponse was proportional to dR/dT for fmod=1 MHz and 4 GHz. The fast response is limited by the electron‐phonon scattering time, estimated to 1.8 ps from experimental data. The responsivity both at 0.8 and 10.6 μm wavelength was ∼1.2 V/W at fmod=1 GHz and the noise equivalent power was calculated to 1.5×10−9 WHz−1/2 for the fast response.
|
|
|
|
Karasik, B. S., Zorin, M. A., Milostnaya, I. I., Elantev, A. I., Gol’tsman, G. N., & Gershenzon, E. M. (1995). Subnanosecond switching of YBaCuO films between superconducting and normal states induced by current pulse. J. Appl. Phys., 77(8), 4064–4070.
Abstract: A study is reported of the current switching in high‐quality YBaCuO films deposited onto NdGaO3 and ZrO2 substrates between superconducting (S) and normal (N) states. The films 60–120 nm thick prepared by laser ablation were structured into single strips between gold contacts. The time dependence of the resistance after application of the voltage step to the film was monitored. Experiment performed within certain ranges of voltage amplitudes and temperatures has shown the occurrence of the fast stage (shorter than 400 ps) both in S‐N and N‐S transitions. A fraction of the film resistance changing within this stage in the S‐N transition increases with the current amplitude. A subnanosecond N‐S stage becomes more pronounced for shorter pulses. The fast switching is followed by the much slower change of resistance. The mechanism of switching is discussed in terms of the hot‐electron phenomena in YBaCuO. The contributions of other thermal processes (e.g., a phonon escape from the film, a heat diffusion in the film and substrate, a resistive domain formation) in the subsequent stage of the resistance dynamic have been also discussed. The basic limiting characteristics (average dissipated power, energy needed for switching, maximum repetition rate) of a picosecond switch which is proposed to be developed are estimated.
|
|
|
|
Kawamura, J., Blundell, R., Tong, C. ‐yu E., Gol’tsman, G., Gershenzon, E., & Voronov, B. (1996). Performance of NbN lattice‐cooled hot‐electron bolometric mixers. J. Appl. Phys., 80(7), 4232–4234.
Abstract: The heterodyne performance of lattice‐cooled hot‐electron bolometric mixers is measured at 200 GHz. Superconducting thin‐film niobium nitride strips with ∼5 nm thickness are used as waveguide mixer elements. A double‐sideband receiver noise temperature of 750 K at 244 GHz is measured at an intermediate frequency centered at 1.5 GHz with 500 MHz bandwidth and with 4.2 K device temperature. The instantaneous bandwidth for this mixer is 1.6 GHz. The local oscillator power required by the mixer is about 0.5 μW. The mixer is linear to within 1 dB up to an input power level 6 dB below the local oscillator power. A receiver incorporating a hot‐electron bolometric mixer was used to detect molecular line emission in a laboratory gascell. This experiment unambiguously confirms that the receiver noise temperature determined from Y‐factor measurements reflects the true heterodyne sensitivity.
|
|
|
|
Semenov, A. D., Hübers, H. - W., Schubert, J., Gol'tsman, G. N., Elantiev, A. I., Voronov, B. M., et al. (2000). Design and performance of the lattice-cooled hot-electron terahertz mixer. J. Appl. Phys., 88(11), 6758–6767.
Abstract: We present the measurements and the theoreticalmodel of the frequency-dependent noise temperature of a superconductor lattice-cooled hot-electron bolometer mixer in the terahertz frequency range. The increase of the noise temperature with frequency is a cumulative effect of the nonuniform distribution of the high-frequency current in the bolometer and the charge imbalance, which occurs at the edges of the normal domain and at the contacts with normal metal. We show that under optimal operation the fluctuation sensitivity of the mixer is determined by thermodynamic fluctuations of the noise power, whereas at small biases there appears additional noise, which is probably due to the flux flow. We propose the prescription of how to minimize the influence of the current distribution on the mixer performance.
|
|
|
|
Semenov, A. D., & Gol’tsman, G. N. (2000). Nonthermal mixing mechanism in a diffusion-cooled hot-electron detector. J. Appl. Phys., 87(1), 502–510.
Abstract: We present an analysis of a diffusion-cooled hot-electron detector fabricated from clean superconducting material with low transition temperature. The distinctive feature of a clean material, i.e., material with large electron mean free path, is a relatively weak inelastic electron scattering that is not sufficient for the establishment of an elevated thermodynamic electron temperature when the detector is subjected to irradiation. We propose an athermal model of a diffusion-cooled detector that relies on suppression of the superconducting energy gap by the actual dynamic distribution of excess quasiparticles. The resistive state of the device is caused by the electric field penetrating into the superconducting bridge from metal contacts. The dependence of the penetration length on the energy gap delivers the detection mechanism. The sources of the electric noise are equilibrium fluctuations of the number of thermal quasiparticles and frequency dependent shot noise. Using material parameters typical for A1, we evaluate performance of the device in the heterodyne regime at terahertz frequencies. Estimates show that the mixer may have a noise temperature of a few quantum limits and a bandwidth of a few tens of GHz, while the required local oscillator power is in the μW range due to ineffective suppression of the energy gap by quasiparticles with high energies.
|
|
|
|
Semenov, A., Engel, A., Il'in, K., Gol'tsman, G., Siegel, M., & Hübers, H. - W. (2003). Ultimate performance of a superconducting quantum detector. Eur. Phys. J. Appl. Phys., 21(3), 171–178.
Abstract: We analyze the ultimate performance of a superconducting quantum detector in order to meet requirements for applications in near-infrared astronomy and X-ray spectroscopy. The detector exploits a combined detection mechanism, in which avalanche quasiparticle multiplication and the supercurrent jointly produce a voltage response to a single absorbed photon via successive formation of a photon-induced and a current-induced normal hotspot in a narrow superconducting strip. The response time of the detector should increase with the photon energy providing energy resolution. Depending on the superconducting material and operation conditions, the cut-off wavelength for the single-photon detection regime varies from infrared waves to visible light. We simulated the performance of the background-limited infrared direct detector and X-ray photon counter utilizing the above mechanism. Low dark count rate and intrinsic low-frequency cut-off allow for realizing a background limited noise equivalent power of 10−20 W Hz−1/2 for a far-infrared direct detector exposed to 4-K background radiation. At low temperatures, the intrinsic response time of the counter is rather determined by diffusion of nonequilibrium electrons than by the rate of energy transfer to phonons. Therefore, thermal fluctuations do not hamper energy resolution of the X-ray photon counter that should be better than 10−3 for 6-keV photons. Comparison of new data obtained with a Nb based detector and previously reported results on NbN quantum detectors support our estimates of ultimate detector performance.
|
|
|
|
Baselmans, J. J. A., Baryshev, A., Reker, S. F., Hajenius, M., Gao, J. R., Klapwijk, T. M., et al. (2006). Influence of the direct response on the heterodyne sensitivity of hot electron bolometer mixers. J. Appl. Phys., 100(8), 084510 (1 to 7).
Abstract: We present a detailed experimental study of the direct detection effect in a small volume (0.15μm×1μm×3.5nm) quasioptical NbN phonon cooled hot electron bolometer mixer at 673GHz. We find that the small signal noise temperature, relevant for an astronomical observation, is 20% lower than the noise temperature obtained using 300 and 77K calibration loads. In a separate set of experiments we show that the direct detection effect is caused by a combination of bias current reduction when switching from the 77 to the 300K
load in combination with the bias current dependence of the receiver gain. The bias current dependence of the receiver gain is shown to be mainly caused by the current dependence of the mixer gain.
|
|
