Search

Large language models (LLMs) have demonstrated their potential in social science research by emulating human perceptions and behaviors, a concept referred to as algorithmic fidelity. This study assesses the algorithmic fidelity and bias of LLMs by utilizing two nationally representative climate change surveys. The LLMs were conditioned on demographics and/or psychological covariates to simulate survey responses. The findings indicate that LLMs can effectively capture presidential voting behaviors but encounter challenges in accurately representing global warming perspectives when relevant covariates are not included. GPT-4 exhibits improved performance when conditioned on both demographics and covariates. However, disparities emerge in LLM estimations of the views of certain groups, with LLMs tending to underestimate worry about global warming among Black Americans. While highlighting the potential of LLMs to aid social science research, these results underscore the importance of meticulous conditioning, model selection, survey question format, and bias assessment when employing LLMs for survey simulation. Further investigation into prompt engineering and algorithm auditing is essential to harness the power of LLMs while addressing their inherent limitations., Comment: 34 pages, 6 figures, 1 table

A novel method utilizing Fast Neutron Resonance Transmission Radiography is proposed for rapid, non-destructive and quantitative determination of the weight fractions of oil and water in cores taken from subterranean or underwater geological formations. Its ability to distinguish water from oil stems from the unambiguously-specific energy-dependence of the neutron cross-sections for the principal elemental constituents. Furthermore, the fluid weight fractions permit determining core porosity and oil and water saturations. In this article we show results of experimental determination of oil and water weight fractions in 10 cm thick samples of Berea Sandstone and Indiana Limestone formations, followed by calculation of their porosity and fluid saturations. The technique may ultimately permit rapid, accurate and non-destructive evaluation of relevant petro-physical properties in thick intact cores. It is suitable for all types of formations including tight shales, clays and oil sands., Comment: 11 pages, 7 figures. arXiv admin note: text overlap with arXiv:1602.06533

A novel method based on Fast Neutron Resonance Transmission Radiography is proposed for non-destructive, quantitative determination of the weight percentages of oil and water in cores taken from subterranean or underwater geological formations. The ability of the method to distinguish water from oil stems from the unambiguously-specific energy dependence of the neutron cross-sections for the principal elemental constituents. Monte-Carlo simulations and initial results of experimental investigations indicate that the technique may provide a rapid, accurate and non-destructive method for quantitative evaluation of core fluids in thick intact cores, including those of tight shales for which the use of conventional core analytical approaches appears to be questionable., Comment: 16 pages, 10 figures, patent pending

The Time-Resolved Integrative Optical Neutron (TRION) detector was developed for Fast Neutron Resonance Radiography (FNRR), a fast-neutron transmission imaging method that exploits characteristic energy-variations of the total scattering cross-section in the En = 1-10 MeV range to detect specific elements within a radiographed object. As opposed to classical event-counting time of flight (ECTOF), it integrates the detector signal during a well-defined neutron Time of Flight window corresponding to a pre-selected energy bin, e.g., the energy-interval spanning a cross-section resonance of an element such as C, O and N. The integrative characteristic of the detector permits loss-free operation at very intense, pulsed neutron fluxes, at a cost however, of recorded temporal resolution degradation. This work presents a theoretical and experimental evaluation of detector related parameters which affect temporal resolution of the TRION system., Comment: Pre-print, also appears at JINST KEYWORDS: Fast neutron detectors; Plastic scintillator; Neutron radiography; Time-of-flight spectroscopy; Detection of explosives; Image intensifier; Pulsed Fast Neutron Transmission Spectroscopy (PFNTS); Fast Neutron Resonance Radiography (FNRR); Simulation

An air cargo inspection system combining two nuclear reaction based techniques, namely Fast-Neutron Resonance Radiography and Dual-Discrete-Energy Gamma Radiography is currently being developed. This system is expected to allow detection of standard and improvised explosives as well as special nuclear materials. An important aspect for the applicability of nuclear techniques in an airport inspection facility is the inventory and lifetimes of radioactive isotopes produced by the neutron and gamma radiation inside the cargo, as well as the dose delivered by these isotopes to people in contact with the cargo during and following the interrogation procedure. Using MCNPX and CINDER90 we have calculated the activation levels for several typical inspection scenarios. One example is the activation of various metal samples embedded in a cotton-filled container. To validate the simulation results, a benchmark experiment was performed, in which metal samples were activated by fast-neutrons in a water-filled glass jar. The induced activity was determined by analyzing the gamma spectra. Based on the calculated radioactive inventory in the container, the dose levels due to the induced gamma radiation were calculated at several distances from the container and in relevant time windows after the irradiation, in order to evaluate the radiation exposure of the cargo handling staff, air crew and passengers during flight. The possibility of remanent long-lived radioactive inventory after cargo is delivered to the client is also of concern and was evaluated., Comment: Proceedings of FNDA 2011

The development of a mm-spatial-resolution, resonant-response detector based on a micrometric glass capillary array filled with liquid scintillator is described. This detector was developed for Gamma Resonance Absorption (GRA) in 14N. GRA is an automatic-decision radiographic screening technique that combines high radiation penetration (the probe is a 9.17 MeV gamma ray) with very good sensitivity and specificity to nitrogenous explosives. Detailed simulation of the detector response to electrons and protons generated by the 9.17 MeV gamma-rays was followed by a proof-of-principle experiment, using a mixed gamma-ray and neutron source. Towards this, a prototype capillary detector was assembled, including the associated filling and readout systems. Simulations and experimental results indeed show that proton tracks are distinguishable from electron tracks at relevant energies, on the basis of a criterion that combines track length and light intensity per unit length., Comment: 18 pages, 16 figures

A stable reference light source based on an LED (Light Emission Diode) is presented for stabilizing the conversion gain of the opto-electronic system of a gamma- and fast-neutron radiographic and tomographic imaging device. A constant fraction of the LED light is transported to the image plane of the camera and provides a stable reference exposure. This is used to normalize the images during off-line image processing. We have investigated parameters influencing the stability of LEDs and developed procedures and criteria to prepare and select LEDs suitable for delivering stable light outputs for several 100 h of operation.

A dual-purpose ion-accelerator concept, capable of serving as radiation source in a versatile, nuclear-reaction-based cargo inspection system, is presented. The system will automatically and reliably detect small, operationally-relevant quantities of concealed explosives and special nuclear materials (SNM). It will be cost-effective, employing largely-common hardware, but different reactions/DAQ-modes. Typical expected throughput is 10-20 aviation containers/hr. PACS: 25.20.Dc; 25.40.Ny; 27.20.+n; 29.27.Fh; 79.77.+g; 89.20.Bb; 89.20.Dd Keywords: Cargo inspection; Nuclear-reaction-based methods; Explosives detection; SNM detection

Two generations of a novel detector for high-resolution transmission imaging and spectrometry of fast-neutrons are presented. These devices are based on a hydrogenous fiber scintillator screen and single- or multiple-gated intensified camera systems (ICCD). This detector is designed for energy-selective neutron radiography with nanosecond-pulsed broad-energy (1 - 10 MeV) neutron beams. Utilizing the Time-of-Flight (TOF) method, such a detector is capable of simultaneously capturing several images, each at a different neutron energy (TOF). In addition, a gamma-ray image can also be simultaneously registered, allowing combined neutron/gamma inspection of objects. This permits combining the sensitivity of the fast-neutron resonance method to low-Z elements with that of gamma radiography to high-Z materials., Comment: Also published in JINST: http://www.iop.org/EJ/abstract/1748-0221/4/05/P05016

TRION is a sub-mm spatial resolution fast neutron imaging detector, which employs an integrative optical time-of-flight technique. The detector was developed for fast neutron resonance radiography, a method capable of detecting a broad range of conventional and improvised explosives. In this study we have analyzed in detail, using Monte-Carlo calculations and experimentally determined parameters, all the processes that influence the signal and noise in the TRION detector. In contrast to event-counting detectors where the signal-to-noise ratio is dependent only on the number of detected events (quantum noise), in an energy-integrating detector additional factors, such as the fluctuations in imparted energy, number of photoelectrons, system gain and other factors will contribute to the noise. The excess noise factor (over the quantum noise) due to these processes was 4.3, 2.7, 2.1, 1.9 and 1.9 for incident neutron energies of 2, 4, 7.5, 10 and 14 MeV, respectively. It is shown that, even under ideal light collection conditions, a fast neutron detection system operating in an integrative mode cannot be quantum-noise-limited due to the relatively large variance in the imparted proton energy and the resulting scintillation light distributions., Comment: 18 pages

In dimension $n>3$ we show the existence of a compactly supported potential in the differentiability class $C^\alpha$, $\alpha < \frac{n-3}2$, for which the solutions to the linear Schr\"odinger equation in $\R^n$, $$ -i\partial_t u = - \Delta u + Vu, \quad u(0)=f, $$ do not obey the usual $L^1\to L^{\infty}$ dispersive estimate. This contrasts with known results in dimensions $n \leq 3$, where a pointwise decay condition on $V$ is generally sufficient to imply dispersive bounds.

We have analyzed and compared the performance of two novel fast-neutron imaging methods with time-of-flight spectroscopy capability. Using MCNP and GEANT code simulations of neutron and charged-particle transport in the detectors, key parameters such as detection efficiency, the amount of energy deposited in the converter and the spatial resolution of both detector variants have been evaluated.

One of our two methods for fast-neutron imaging with spectrometric capability is presented here. It is a neutron-counting technique based on a hydrogenous neutron converter coupled to Gaseous Electron Multipliers (GEM). The principles of the detection techniques and the optimization of the converter, electron amplification and the readout are described. Evaluation of the properties are derived from a experiment in a pulsed neutron beam of spectral distribution between 2 and 10 MeV

Two detectors for energy-resolved fast-neutron imaging in pulsed broad-energy neutron beams are presented. The first one is a neutron-counting detector based on a solid neutron converter coupled to a gaseous electron multiplier (GEM). The second is an integrating imaging technique, based on a scintillator for neutron conversion and an optical imaging system with fast framing capability.

We prove L^1 --> L^\infty estimates for linear Schroedinger equations in dimensions one and three. The potentials are only required to satisfy some mild decay assumptions. No regularity on the potentials is assumed., Comment: 20 pages. Corrected typos and improved explanatory remarks at the end