Noirot J.,CEA Cadarache Center |
Lamontagne J.,CEA Cadarache Center |
Nakae N.,JNES |
Kitagawa T.,MNF |
Journal of Nuclear Materials | Year: 2013
A UO2 fuel with a heterogeneous distribution of 235U was irradiated up to a high burn-up in the Halden Boiling Water Reactor (HBWR). The last 100 days of irradiation were performed with an increased level of linear power. The effect of the heterogeneous fissile isotope distribution on the formation of the HBS was studied free of the possible influence of Pu which exists in heterogeneous MOX fuels. The HBS formed in 235U-rich agglomerates and its main characteristics were very similar to those of the HBS formed in Pu-rich agglomerates of heterogeneous MOX fuels. The maximum local contents of Nd and Xe before HBS formation were studied in this fuel. In addition to a Pu effect that promotes the HBS phenomenon, comparison with previous results for heterogeneous MOX fuels showed that the local fission product concentration was not the only parameter that has to be taken into consideration. It appears that the local actinide depletion by fission and/or the energy locally deposited through electronic interactions in the fission fragment recoils also have an effect on the HBS formation threshold. Moreover, a major release of fission gases from the peripheral 235U-rich agglomerates of HBS bubbles and a Cs radial movement are also evidenced in this heterogeneous UO2. Cs deposits on the peripheral grain boundaries, including the HBS grain boundaries, are considered to reveal the release paths. © 2013 Elsevier B.V. All rights reserved.
Ye J.,University of California at Berkeley |
Ye J.,Lawrence Berkeley National Laboratory |
Mishra R.K.,General Motors |
Pelton A.R.,NDC |
And 2 more authors.
Acta Materialia | Year: 2010
We report on quantitative in situ transmission electron microscopy nanocompression tests used to study the deformation behavior of NiTi pillars on the nanometer scale. By recording the diffraction patterns in real time we have obtained direct evidence that the stress-induced B2 to B19′ (austenite to martensite) transformation exists in NiTi even when the sample size is below 200 nm. Correlation of the appearance of the B19′ phase in the diffraction pattern with our quantitative data showed that the transformation starts at approximately 1 GPa. We found that the transformation occurred through a multi-step process, and that the reverse transformation did not occur due to extensive deformation of the B19′ phase. Our results have direct implications for the application of the shape memory effect in nanoscale NiTi devices.
News Article | April 8, 2016
Papua New Guinea recently became the first country to formally submit the final version of its national climate action plan (called a “Nationally Determined Contribution,” or NDC) under the Paris Agreement. The small Pacific nation’s plan to transition to 100 percent renewable energy by 2030 is no longer just an “intended” nationally determined contribution (INDC) — it is now the country’s official climate plan.
The set of structures we investigate consists of InAs nanowires grown by molecular beam epitaxy in the  wurtzite direction with an epitaxial aluminium (Al) shell on two facets of the hexagonal cross-section18. The Al shell was removed except in a small segment of length L and isolated from normal metal (titanium/gold) leads by electrostatic gate-controlled barriers (Fig. 1a). The charging energies E of the measured devices range from greater than to less than the superconducting gap of Al (approximately 0.2 meV). The thinness of the Al shell (8–10 nm on the two facets) results in a large critical field B before superconductivity is destroyed: for fields along the wire axis, B ≈ 1 T; out of the plane of the substrate, but roughly in the plane of the two Al-covered facets, B ≈ 700 mT (Fig. 1b). The very high critical fields that are achieved make these wires a suitable platform for investigating topological superconductivity18. Five devices over a range of Al shell lengths L ≈ 0.3–1.5 μm were measured (see Methods for device layouts). Charge occupation and tunnel coupling to the leads were tuned via electrostatic gates. Differential conductance g in the Coulomb-blockade regime (high-resistance barriers) was measured using standard a.c. lock-in techniques in a dilution refrigerator (electron temperature of about 50 mK). Figure 1c shows g as a function of gate voltage V and source–drain bias V . For the L = 790 nm device, the zero-field data (Fig. 1c, top) show a series of evenly spaced Coulomb diamonds with a characteristic negative-differential conductance (NDC) region at higher bias. NDC is known from metallic superconductor islands19, 20 and has recently been reported in a proximitized semiconductor device similar to those investigated here21. The zero-magnetic-field diamonds reflect charge transport via Cooper pairs, with gate-voltage period proportional to 2e, the charge of a Cooper pair. At moderate magnetic fields (Fig. 1c, middle), the large diamonds shrink and a second set of diamonds appears, yielding even–odd spacing of Coulomb-blockade zero-bias conductance peaks22, as seen in the bottom panel of Fig. 1d. At larger magnetic fields (Fig. 1c, bottom), Coulomb diamonds are again periodic, but have precisely half the spacing of the zero-field diamonds, corresponding to 1e periodicity. At this field NDC is absent, and resonant structure is visible within each diamond, indicating transport through discrete resonances at low bias and a continuum at high bias (see magnification in Fig. 1c). Coulomb-blockade conductance peaks at high magnetic field (see Fig. 1d for zero-bias cross-sections) with regular 1e periodicity (half the zero-field spacing) accompanied by a discrete sub-gap spectrum are a proposed signature of electron teleportation by Majorana end states16, 17. We designate the ungrounded tunnelling device in this high-field regime as a ‘Majorana island’, where a sub-gap state near zero energy, energetically isolated from a continuum, leads to 1e-periodic Coulomb-blockade conductance peaks. Zero-bias conductance can be qualitatively understood in a simple zero-temperature model in which the energy of the superconducting island—with or without sub-gap states (Fig. 1d)—is given by a series of shifted parabolas: E (N ) = E (N −N)2 + p E , in which N = CV /e is the gate-induced charge (with electron charge e and gate capacitance C)19, 20, 22, 23, 24, 25 and N is the electron occupancy. E is the energy of the lowest quasiparticle state, which is filled for odd parity (p = 1, odd N) and empty for even parity (p = 0, even N)21. Transport occurs when the ground state has a charge degeneracy, that is, when the E parabolas intersect. For E > E , the ground state always has even parity; transport in this regime occurs via tunnelling of Cooper pairs at degeneracies of the even-N parabolas. This is the regime in which the 2e-periodic Coulomb-blockade peaks are seen at low magnetic fields (Fig. 1d, blue). The odd charge state carries spin and its energy can be lowered by the Zeeman effect when a magnetic field is applied. For sufficiently large field, such that E < E , an odd-N ground state emerges. This transition from 2e charging to 1e charging is seen experimentally as the splitting of the 2e-periodic Coulomb diamonds into the even–odd double-diamond pattern in Fig. 1d (green). In this regime, the Coulomb-peak spacing is proportional to E + 2E for even diamonds and E −2E for odd diamonds23, 24. For the particular case of a zero-energy Majorana state (E = 0) peak spacing is regular and 1e-periodic. This regime is observed at higher fields (Fig. 1d, red), although not so high as to destroy superconductivity. Coulomb-peak spacings are measured as a function of magnetic field, allowing the state energy, E (B), to be extracted. An example, showing ten consecutive peaks for the L = 0.9 μm device, is shown in Fig. 2a. The peaks are 2e-periodic at B = 0, start splitting at B ≈ 95 mT and become 1e-periodic at B ≈ 110 mT, well below the spectroscopically observed closing of the superconducting gap at B ≈ 600 mT (see Methods). This result indicates the presence of a state close to zero energy within the superconducting regime over a range of about 500 mT. Separately averaging even and odd Coulomb-peak spacings (〈S 〉) over an ensemble of adjacent peaks reveals oscillations around the 1e-periodic value as a function of applied magnetic field. This finding is consistent with an oscillating state energy E due to hybridized Majorana modes13, 14, 15. For the L = 0.9 μm device (Fig. 2b), peak-spacing oscillations yield an energy oscillation amplitude A = 7.0 ± 1.5 μeV that is converted from gate voltage to energy using the gate lever arm η, which is extracted independently from the slope of the Coulomb diamonds. For the L = 1.5 μm device (Fig. 2c), oscillations in the average Coulomb-peak spacing determined from 22 consecutive peaks yield a barely resolvable amplitude A = 1.2 ± 0.5 μeV. Oscillation amplitudes for the five measured devices (see Methods for device details) are shown in Fig. 2d along with a two-parameter fit to an exponential function, A = A e−L/ξ, which yields A = 300 μeV and ξ = 260 nm as fit parameters. The data fit well to the predicted exponential form that characterizes the topological protection of Majorana modes3, 4, 13. Excited states of the Majorana island are probed using finite-bias transport spectroscopy. This technique requires a fixed gate voltage, which is chosen such that, at zero bias, the electrochemical potential of the leads aligns with the centre of the spectroscopic gap of the Majorana island. With this choice, the conductance observed at a source–drain bias V is due to states at energy eV /2. A conductance peak at zero bias corresponds to a zero-energy state. In the case shown in Fig. 3a, b, the gate voltage is tuned using the characteristic finite-bias conductance spectra for a short InAs/Al island that was investigated previously21. Ground-state energies determined by finite-bias spectroscopy match those extracted from zero-bias peak spacings (see Extended Data Fig. 7). Bias spectroscopy shows discrete zero-energy states emerging at sufficient applied field over a range of device lengths. In a short device (Fig. 3c), the discrete state moves linearly as a function of magnetic field, passing through zero and merging with a continuum at V ≈ 100 μeV. This merging is expected for Majorana systems in the short-length limit, in which quenching of spin–orbit coupling results in unprotected parity crossings and state intersections at high energy14. However, rather than passing directly through zero, the first zero crossing extends for 40 mT; this behaviour is not understood. For medium-length devices, the sub-gap state bends back towards zero after zero crossings (Fig. 3d), in agreement with theoretical predictions for the emergence of Majorana behaviour with increasing system length14, 15. For a long device (L = 1.5 μm), bias spectroscopy shows a zero-energy state separated from a continuum at higher bias (Fig. 3e). The zero-energy state is present over a field range of 120 mT, with an associated energy gap of (30 μeV)/k = 0.35 K (in which k is the Boltzmann constant). The evolution with increasing device length from unprotected parity crossings to energetically isolated oscillating states and then to a fixed zero-energy state is consistent with the expected crossover from a strongly overlapping precursor of split Majorana states to a topologically protected Majorana state locked at zero energy14, 15. In the data in Fig. 3e, the signal from the discrete state disappears for B > 320 mT. This is not expected for a simple (disorder-free, single sub-band) Majorana picture. Even though the zero-bias peak disappears, the peak spacing remains 1e-periodic (see Methods). The observed effective g-factors of 20–50, which are extracted from the addition spectrum and bias spectroscopy (see Methods), are large compared to previous studies on InAs nanowires9, 26, 27, perhaps as a result of field focusing from the Al shell. The measured gap to the continuum at zero magnetic field is consistent with the gap of aluminium, Δ ≈ 180 μeV, and is roughly the same in all devices. The discrete subgap states (Fig. 3c–e) have zero-field energy that is less than, but comparable to, the gap, E (B = 0) ≈ 50–160 μeV, which is consistent with expectations for half-shell geometries28. The measured gap between the near-zero-energy state and the continuum in the high-field (topological) regime, Δ ≈ 30 μeV, as well as the coherence length extracted from the exponential fit to the length-dependent splitting (Fig. 2d), ξ ≈ 260 nm, are consistent with topological superconductivity. At low magnetic fields, the gap and coherence length are related to the strength of spin–orbit coupling: α ≈ ξ × Δ = 8 × 10−2 eV Å; this value is consistent with those previously reported for InAs nanowires9, 29. For a single sub-band, this implies a Fermi velocity v = α /ħ = 1 × 104 m s−1 that is lower than expected, suggesting that more than one sub-band is occupied under the Al shell; however, we are not able to extract the number of modes directly. Finally, we consider the magnetic-field dependence of the heights of Coulomb-blockade peaks (as opposed to the spacings) (Fig. 4). We found in most devices that below the field B*, at which 2e-periodic peaks split, all peaks had uniformly high amplitude. Above B*, peak heights rapidly decreased and remained low up to a second characteristic field, B**, coincident with the onset of 1e periodicity (that is, the field at which even–odd spacing differences vanished). Above B**, peak heights recovered. In the longer wires, peaks were nearly absent between B* and B** (Fig. 4c). We interpret these observations as follows. In the present lead–wire–lead geometry, transport at fields above B* involves single electrons entering one end of the wire and leaving from the other. The onset of uniform spacing with the reappearance of high peaks for fields above B** indicates the emergence of a state (or states) at zero energy with strong wavefunction weight at both ends of the wire. This is consistent with teleportation of electrons from one end of the wire to the other via a Majorana mode16, 17, although it is not necessarily a unique signature of teleportation30. Therefore, although the simultaneous brightening of peaks and their becoming uniformly spaced at B** suggests a sub-gap or Majorana mode moving to the ends of the wire as it moves to zero energy, we cannot rule out other forms of end-localized zero-energy states that could appear above a critical field. In summary, we studied Majorana islands composed of InAs nanowires covered on two facets with epitaxial Al, for a range of device lengths. Zero-energy states are observed for wires of all lengths away from zero field. Oscillating energy splittings, measured using Coulomb-blockade spectroscopy, are exponentially suppressed with wire length, with a characteristic length ξ = 260 nm. This result constitutes an explicit demonstration of exponential protection of zero-energy modes. Finite-bias measurements show transport through a discrete zero-energy state, with a measured topological gap Δ = 30 μeV for long devices. The extracted Δ and ξ are consistent with known parameters for InAs nanowires and the emergence of topological superconductivity. Brightening of Coulomb peaks at the field at which spacing becomes uniform for longer devices suggests the presence of a robust delocalized state connecting the leads, and provides experimental support for electron teleportation via Majorana modes.
Goyal R.,Software Group |
Jayasudha T.,Software Group |
Pandey P.,CBIT |
Devi D.R.,NDC |
And 3 more authors.
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives | Year: 2014
In recent years, the use of satellite data for geospatial applications has multiplied and contributed significantly towards development of the society. Satellite data requirements, in terms of spatial and spectral resolution, periodicity of data, level of correction and other parameters, vary for different applications. For major applications, remote sensing data alone may not suffice and may require additional data like field data. An application user, even though being versatile in his application, may not know which satellite data is best suited for his application, how to use the data and what information can be derived from the data. Remote sensing domain experts have the proficiency of using appropriate data for remote sensing applications. Entrenching domain expertise into the system and building a knowledge base system for satellite data product selection is vital. Non specialist data users need a user-friendly software which guides them to the most suitable satellite data product on the basis of their application. Such tool will aid the usage for apt remote sensed data for various sectors of application users. Additionally, the consumers will be less concerned about the technical particulars of the platforms that provide satellite data, instead focusing on the content and values in the data product, meeting the timelines and ease of access. Embedding knowledge is a popular and effective means of increasing the power of using a system. This paper describes a system, driven by the built-in knowledge of domain experts, for satellite data products selection for geospatial applications.