Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Power Law”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
NASA NTRS2019

An Evolving Broad Iron Line from the First Galactic Ultraluminous X-Ray Pulsar Swift J0243.6+6124

We present a spectral study of the ultraluminous Be/X-ray transient pulsar Swift J0243.6+6124 using Neutron Star Interior Composition Explorer (NICER) observations during the system's 2017–2018 giant outburst. The 1.2–10 keV energy spectrum of the source can be approximated with an absorbed cutoff power-law model. We detect strong, luminosity-dependent emission lines in the 6–7 keV energy range. A narrow 6.42 keV line, observed in the sub-Eddington regime, is seen to evolve into a broad Fe-line profile in the super-Eddington regime. Other features are found at 6.67 and 6.97 keV in the Fe-line complex. An asymmetric broad-line profile, peaking at 6.67 keV, is possibly due to Doppler effects and gravitational redshift. The 1.2–79 keV broadband spectrum from Nuclear Spectroscopic Telescope Array (NuSTAR) and NICER observations at the outburst peak is well described by an absorbed cutoff power law plus multiple Gaussian lines and a blackbody component. Physical reflection models are also tested to probe the broad iron-line feature. Depending on the mass accretion rate, we found emission sites that are evolving from ~5000 km to a range closer to the surface of the neutron star. Our findings are discussed in the framework of the accretion disk and its implication on the magnetic field, the presence of optically thick accretion curtain in the magnetosphere, jet emission, and the massive, ultrafast outflow expected at super-Eddington accretion rates. We do not detect any signatures of a cyclotron absorption line in the NICER or NuSTAR data.

Gaurava K. Jaisawal↗
NASA NTRS1989

Self-similar Lagrangian hydrodynamics of beam-heated solar flare atmospheres

The one-dimensional hydrodynamic problem in Lagrangian coordinates (Y, t) is considered for which the specific energy input Q has a power-law dependence on both Y and t, and the initial density distribution is rho(0) which is directly proportional to Y exp gamma. In regimes where the contributions of radiation, conduction, quiescent heating, and gravitational terms in the energy equation are negligible compared to those arising from Q, the problem has a self-similar solution, with the hydrodynamic variables depending only on a single independent variable which is a combination of Y, t, and the dimensional constants of the problem. It is then shown that the problem of solar flare chromospheric heating due to collisional interaction of a beam of electrons (or protons) with a power-law energy spectrum can be approximated by such forms of Q(Y, t) and rho(0)(Y), and that other terms are negligible compared to Q over a restricted regime early in the flare.

Brown, John C.↗
DOE OSTI2023

Aperture size distribution, length, and preferential location of bed-parallel veins in shale

Bed-parallel, calcite-filled veins (BPVs) are common in shale formations, and although they have been widely described in other studies, little is known about their population aperture size distribution. To address this knowledge gap, we analyzed BPV sizes in outcrops and cores from the Vaca Muerta Formation, Neuquén Basin, Argentina; in two cores from the Marcellus Formation, Appalachian Basin, northeast Pennsylvania; and in one core from the Wolfcamp Shale, Delaware Basin, West Texas. Nine out of ten aperture size populations follow a negative exponential distribution, with one following a weak power law. Bed-parallel vein size distribution and intensity vary among formations and within the same shale. We define three groups of distributions: (1) Vaca Muerta outcrops, with the highest BPV intensity and the largest BPVs (cumulative frequency of 4.9 BPVs per meter [BPVs/m] for apertures 0.265 mm to 8.7 cm); (2) Vaca Muerta cores with a similar BPV intensity overall but with no apertures wider than 1.2 cm; and (3) Vaca Muerta, Wolfcamp, and Marcellus cores with the fewest BPVs (cumulative frequency up to 0.63 BPVs/m) and very few wider than 1 cm. Aperture and length in two outcrop data sets are weakly positively correlated and follow power laws with exponents of 0.44 and 0.49. Mechanical interfaces at boundaries between different lithologies exert a strong control on BPV location, with 65–75% of observed interfaces having BPVs along them. Only 25–30% of the BPVs occur at observed material interfaces, however, and unless subtle, unobserved mechanical layering is present, other factors must also control location. BPV intensity and organic richness (TOC) from Vaca Muerta well logs are correlated in some instances but not in others, indicating TOC is not always a good proxy for BPV location or intensity. Furthermore, these findings provide useful information for modeling of hydraulic fracture treatments where BPVs may influence development of the stimulated fracture network, for example by limiting height growth.

58 GEOSCIENCES↗
NASA NTRS1994

X-ray emission associated with radio galaxies in the Perseus cluster

In this paper, we report on new x-ray observations of the Perseus cluster made using four separate pointings of the Roentgen Satellite (ROSAT) Positron Sensitive Proportional Counter (PSPC). We searched for x-ray emission associated with 16 radio galaxies and detected six above 3 sigma. We made use of the PSPC spectra to determine if the x-ray emission associated with radio galaxies in Perseus is thermal or nonthermal in origin (i.e., hot gas or an active galactic nuclei (AGN)). For the head-tail radio galaxy IC 310, we find that the data are best fit by a power law model with an unusually large spectral index alpha = 2.7. This is consistent with its unresolved spatial structure. On the other hand, a second resolved x-ray source associated with another radio galaxy 2.3 Mpc from the Perseus center (V Zw 331) is best fit by a thermal model. For three sources with insufficient flux for a full spectral analysis, we calculated hardness ratios. On this basis, the x-ray emission associated with the well known head-tail source NGC 1265 is consistent with thermal radiation. The x-ray spectra of UGC 2608 and UGC 2654 probably arise from hot gas, although very steep power-law spectra (alpha greater than 3.2) are also possible. The spectrum of NGC 1275 is quite complex due to the presence of an AGN and the galaxy's location at the center of a cluster cooling flow.

Rhee, George↗
NASA NTRS1992

The broad-band x ray spectral variability of Mkn 841

The results of a detailed spectral analysis of four X-ray observations of the luminous Seyfert 1.5 galaxy Mkn 841 performed using the EXOSAT and Ginga satellites over the period June 1984 to July 1990 are reported. Preliminary results from a short ROSAT PSPC observation of Mkn 841 in July 1990 are also presented. Variability is apparent in both the soft (0.1-1.0 keV) and medium (1-20 keV) energy bands. Above 1 keV, the spectra are adequately modelled by a power-law with a strong emission line of equivalent width approximately 450 eV. The energy of the line (approximately 6.4 keV) is indicative of K-shell fluorescence from neutral iron, leading to the interpretation that the line arises via X-ray illumination of cold material surrounding the source. In addition to the flux variability, the continuum shape also changes in a dramatic fashion, with variations in the apparent photon index Delta(Gamma) approximately 0.6. The large equivalent width of the emission line clearly indicates a strongly enhanced reflection component in the source, compared to other Seyferts observed with Ginga. The spectral changes are interpreted in terms of a variable power-law continuum superimposed on a flatter reflection component. For one Ginga observation, the reflected flux appears to dominate the medium energy X-ray emission, resulting in an unusually flat slope (Gamma approximately 1.0). The soft X-ray excess is found to be highly variable by a factor approximately 10. These variations are not correlated with the hard flux, but it seems likely that the soft component arises via reprocessing of the hard X-rays. We find no evidence for intrinsic absorption, with the equivalent hydrogen column density constrained to be less than or equal to few x 10(exp 20) cm(exp -2). The implications of these results for physical models for the emission regions in this and other X-ray bright Seyferts are briefly discussed.

George, I. M.↗
NASA NTRS2020

The Fast Radio Burst Luminosity Function and Death Line in the Low-Twist Magnetar Model

We explore the burst energy distribution of fast radio bursts(FRBs)in the low-twist magnetar model ofWadiasingh & Timokhin(WT19). Motivated by the power-lawfluence distributions of FRB121102, we proposean elementary model for the FRB luminosity function of individual repeaters with an inversion protocol thatdirectly relates the power-law distribution index of magnetar short burstfluences to that for FRBs. The protocolindicates that the FRB energy scales virtually linearly with crust/field dislocation amplitude, if magnetar shortbursts prevail in the magnetoelastic regime. Charge starvation in the magnetosphere during bursts(required inWT19)for individual repeaters implies the predicted burstfluence distribution is narrow,3 decades for yieldingstrains and oscillation frequencies feasible in magnetar crusts. Requiring magnetic confinement and chargestarvation, we obtain a death line for FRBs, which segregates magnetars from the normal pulsar population,suggesting only the former will host recurrent FRBs. We convolve the burst energy distribution for individualmagnetars to define the distribution of luminosities in evolved magnetar populations. The broken power-lawluminosity function’s low-energy character depends on the population model, while the high-energy index tracesthat of individual repeaters. Independent of the evolved population, the broken power-law isotropic-equivalentenergy/luminosity function peaks at∼1037–1040erg with a low-energy cutoff at∼1037erg. Lastly, we considerthe localfluence distribution of FRBs andfind that it can constrain the subset of FRB-producing magnetarprogenitors. Our model suggests that improvements in sensitivity may reveal aflattening of the global FRBfluencedistribution and saturation in FRB rates.

Cosmology↗
NASA NTRS1985

The electron spectra in the synchrotron nebula of the supernova remnant G 29.7-0.3

EXOSAT results obtained with the imaging instrument (CMA) and the medium energy proportional counters (ME) are discussed. Assuming that the featureless power-law spectrum obtained in the 2 to 10 keV range is synchrotron radiation from relativistic electrons, one derives constraints on magnetic field strength and age of the nebula. The energy spectra of the electrons responsible for the emission in the radio and X-ray ranges are discussed. The great similarity of the physical properties of G 29.7-0.3 and of three synchrotron nebulae containing a compact object observed to pulse in X-rays makes G 29.7 - 0.3 a very promising candidate for further search for pulsed emission. Further observations at infrared wavelengths might reveal the break(s) in the emitted spectrum expected from the radio and X-ray power-law indices and give us more information on the production of the electron populations responsible for the emission of the nebula.

Koch-Miramond, L.↗
NASA NTRS1993

Warm O(+) polar wind and the DE-1 polar cap electron density profile

Theoretical steady state semikinetic polar wind density profiles, based on DE1/RIMS polar wind data (up to 3700 km), were obtained which agree very well with the power law electron density profile measured by the DE1/PWI for high altitudes. The polar wind is found to be O(+) dominated for the full altitude range considered (up to 8 R(E)). Multiple solutions are obtained for various combinations of base altitude ion temperatures and electron temperatures, such that the densities fit the Persoon et al. (1983) profile. For example, good fits to measured density profile are found for low base ion temperatures (5000 K) and high electron temperatures (9000 K), and also for unheated H(+) and O(+)(3000 K) with electron temperatures of 11,000 K. Below 2.8 R(E) the theoretical polar wind density deviates somewhat from the r exp -3.85 power law. It is concluded that this theoretical polar wind density profile, with a sum of base electron and ion temperatures of 14,000 K, yields a close match with the measured DE-1 electron density profile.

Ho, C. W.↗
DOE OSTI2022

Side Chain Dynamics of Poly(norbornene)- g -Poly(propylene oxide) Bottlebrush Polymers

The segmental dynamics of the side chains of poly(norbornene)- g -poly(propylene oxide) (PNB- g -PPO) bottlebrush polymer in comparison to PPO is studied by quasi-elastic neutron scattering. Having experimental time and length scale information simultaneously allows to extract spatial information in addition to relaxation time. Tethering one end of the PPO side chain onto a stiff PNB backbone slows down the segmental relaxation, over the length and time scales investigated. The power law dependence of the relaxation time on the momentum transfer, Q , indicates a more heterogeneous relaxation pattern for the bottlebrush polymer, whereas the linear PPO has less deviations from a homogenous relaxation. Similar conclusions can be drawn from the time dependent mean square displacement, $\langle$ r 2 ( t )$\rangle$, and the non-Gaussian parameter, α 2 ( t ). Herein, the bottlebrush polymer shows a more restricted dynamics, whereas the linear PPO reaches $\langle$ r 2 ( t )$\rangle$∝ t 0.5 at the highest temperature. The deviations from Gaussian behavior are evident at the α 2 ( t ). Both samples show a decaying α 2 ( t ). Finally, the non-Gaussian parameter supports the results from the power law dependence of the relaxation times, with lower α 2 ( t ) values for PPO compared to those for PNB-g-PPO, pointing to less deviations from Gaussian behavior.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Guide field effects on the distribution of plasmoids in multiple scale reconnection

The effects of a finite guide field on the distribution of plasmoids in high-Lundquist-number current sheets undergoing magnetic reconnection in large plasmas are investigated with statistical models. Merging of plasmoids is taken into account either assuming that guide field flux is conserved resulting in nonforce-free profiles in general, or that magnetic helicity is conserved and Taylor relaxation occurs to convert part of the summed guide field flux into reconnecting field flux toward minimum energy states resulting in force-free profiles. It is found that the plasmoid distribution in terms of reconnecting field flux follows a power law with index 7/4 or 1 depending on whether merger frequencies are independent of or dependent on their relative velocity to the outflow speed, respectively. This result is approximately the same for the force-free and nonforce-free models, with nonforce-free models exhibiting indices of 2 and 1 for the same velocity dependencies. Distributions in terms of guide field flux yield indices of 3/2 for the nonforce-free model regardless of velocity dependence. This is notably distinct from the indices of 11/8 and 1 for the force-free models independent of and dependent on velocity, respectively. At low guide field fluxes, the force-free models exhibit a second power law index of 1/2 due to nonconstant flux growth rates. The velocity-dependent force-free model predicts the production of slightly more rapidly moving large guide field flux plasmoids which are supported by observational evidence of flux ropes with strong core fields. Implications are discussed on particle acceleration via Fermi processes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1992

Fractal landscapes in biological systems: long-range correlations in DNA and interbeat heart intervals

Here we discuss recent advances in applying ideas of fractals and disordered systems to two topics of biological interest, both topics having common the appearance of scale-free phenomena, i.e., correlations that have no characteristic length scale, typically exhibited by physical systems near a critical point and dynamical systems far from equilibrium. (i) DNA nucleotide sequences have traditionally been analyzed using models which incorporate the possibility of short-range nucleotide correlations. We found, instead, a remarkably long-range power law correlation. We found such long-range correlations in intron-containing genes and in non-transcribed regulatory DNA sequences as well as intragenomic DNA, but not in cDNA sequences or intron-less genes. We also found that the myosin heavy chain family gene evolution increases the fractal complexity of the DNA landscapes, consistent with the intron-late hypothesis of gene evolution. (ii) The healthy heartbeat is traditionally thought to be regulated according to the classical principle of homeostasis, whereby physiologic systems operate to reduce variability and achieve an equilibrium-like state. We found, however, that under normal conditions, beat-to-beat fluctuations in heart rate display long-range power law correlations.

Non-NASA Center↗
NASA NTRS2006

The Onset Phase of "Soft" X-ray Transients

Transient outbursts of black holes and neutron stars in X-ray binaries with low-mass companions start with a flickering hard power-law flux that contains a low frequency quasi-periodic oscillation (QPO). The frequency of the QPO may reflect the outer boundary of the coronal emission and its inward motion toward the compact object. It has also been proposed that the hard flux is related to the base of a radio emitting outflow or compact jet. We had detailed observations of the beginning of the 2005 outburst of GRO J165.5-40 with RXTE, INTEGRAL, the VLA and ROTSE. We use the X-ray, radio, and optical results in the context of these models to address their applicability to the onset of the outburst and to specify the physical parameters. Decline of the radio flux as both the power-law and disk flux increased constrains the amount of synchrotron self-Compton emission. Values are compared to those of other black hole and neutron star transients. We are glad to acknowledge support by a NASA INTEGRAL Guest Observer Grant and by the UTE project, NRAO, and ROTSE.

Swank, Jean Hebb↗
DOE OSTI2023

Four Years of Type Ia Supernovae Observed by TESS: Early-time Light-curve Shapes and Constraints on Companion Interaction Models

Abstract We present 307 type Ia supernova (SN) light curves from the first 4 yr of the Transiting Exoplanet Survey Satellite mission. We use this sample to characterize the shapes of the early-time light curves, measure the rise times from first light to peak, and search for companion star interactions. Using simulations, we show that light curves must have noise <10% of the peak flux to avoid biases in the early-time light-curve shape, restricting our quantitative analysis to 74 light curves. We find that the mean power-law index t β 1 of the early-time light curves isβ 1 = 1.93 ± 0.57, and the mean rise time to peak is 15.7 ± 3.5 days. The underlying population distribution forβ 1 may instead consist of a Gaussian component with mean 2.29, width 0.34, and a long tail extending to values less than 1.0. We find that the data can rarely distinguish between models with and without companion interaction models. Nevertheless, we find three high-quality light curves that tentatively prefer the addition of a companion interaction model, but the statistical evidence for the companion interactions is not robust. We also find two SNe that disfavor the addition of a companion interaction model to a curved power-law model. Taking the 74 SNe together, we calculate 3σupper limits on the presence of companion signatures to control for orientation effects that can hide companions in individual light curves. Our results rule out common progenitor systems with companions having Roche lobe radii >31R ⊙ (separations >5.7 × 10 12 cm, 99.9% confidence level) and disfavor companions having Roche lobe radii >10R ⊙ (separations >1.9 × 10 12 cm, 95% confidence level). Lastly, we discuss the implications of our results for the intrinsic fraction of single degenerate progenitor systems.

Astronomy & Astrophysics↗
NASA NTRS2012

Understanding the Impact of Return-Current Losses on the X-Ray Emission from Solar Flares

I obtain and examine the implications of one-dimensional analytic solutions for return-current losses on an initially power-law distribution of energetic electrons with a sharp low-energy cutoff in flare plasma with classical (collisional) resistivity. These solutions show, for example, that return-current losses are not sensitive to plasma density, but are sensitive to plasma temperature and the low energy cutoff of the injected nonthermal electron distribution. A characteristic distance from the electron injection site, x(sub rc), is derived. At distances less than x(sub rc) the electron flux density is not reduced by return-current losses, but plasma heating can be substantial in this region, in the upper, coronal part of the flare loop. Before the electrons reach the collisional thick-target region of the flare loop, an injected power-law electron distribution with a low-energy cutoff maintains that structure, but with a flat energy distribution below the cutoff energy, which is now determined by the total potential drop experienced by the electrons. Modifications due to the presence of collisional losses are discussed. I compare these results with earlier analytical results and with more recent numerical simulations. Emslie's 1980 conjecture that there is a maximum integrated X-ray source brightness on the order of 10(exp -15) photons per square centimeter per second per square centimeter is examined. I find that this is not actually a maximum brightness and its value is parameter dependent, but it is nevertheless a valuable benchmark for identifying return-current losses in hard X-ray spectra. I discuss an observational approach to identifying return-current losses in flare data, including identification of a return-current "bump" in X-ray light curves at low photon energies.

Holman, Gordon D.↗
NASA NTRS2017

Studying the Accretion Geometry of EXO 2030+375 at Luminosities Close to the Propeller Regime

The Be X-ray binary EXO2030+375 was in an extended low-luminosity state during most of 2016. We observed this state with NuSTAR and Swift, supported by INTEGRAL observations and optical spectroscopy with the Nordic Optical Telescope (NOT). We present a comprehensive spectral and timing analysis of these data here to study the accretion geometry and investigate a possible onset of the propeller e ect. The H data show that the circumstellar disk of the Be-star is still present. We measure equivalent widths similar to values found during more active phases in the past, indicating that the low-luminosity state is not simply triggered by a smaller Be disk. The NuSTAR data, taken at a 3-78 keV luminosity of 6:8 1035 erg s-1 (for a distance of 7.1 kpc), are nicely described by standard accreting pulsar models such as an absorbed power law with a high-energy cuto. We find that pulsations are still clearly visible at these luminosities, indicating that accretion is continuing despite the very low mass transfer rate. In phaseresolved spectroscopy we find a peculiar variation of the photon index from 1.5 to 2.5 over only about 3% of the rotational period. This variation is similar to that observed with XMM-Newton at much higher luminosities. It may be connected to the accretion column passing through our line of sight. With Swift/XRT we observe luminosities as low as 1034 erg s-1 where the data quality did not allow us to search for pulsations, but the spectrum is much softer and well described by either a blackbody or soft power-law continuum. This softer spectrum might be due to the accretion being stopped by the propeller e ect and we only observe the neutron star surface cooling.

Fürst, F.↗
DOE OSTI2022

On the origin of multihole oxygen evolution in haematite photoanodes

The oxygen evolution reaction (OER) plays a crucial role in (photo)electrochemical devices that use renewable energy to produce synthetic fuels. Recent measurements on semiconducting oxides have found a power law dependence of the OER rate on surface hole density, suggesting a multihole mechanism. In this study, using transient photocurrent measurements, density functional theory simulations and microkinetic modelling, we have uncovered the origin of this behaviour in haematite. We show here that the OER rate has a third-order dependence on the surface hole density. We propose a mechanism wherein the reaction proceeds by accumulating oxidizing equivalents through a sequence of one-electron oxidations of surface hydroxy groups. The key O–O bond formation step occurs by the dissociative chemisorption of a hydroxide ion involving three oxyl sites. At variance with the case of metallic oxides, the activation energy of this step is weakly dependent on the surface hole coverage, leading to the observed power law.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Universal temperature dependence of the London penetration depth in κ - ( ET ) 2 X superconductors

High-precision radio-frequency magnetic susceptibility measurements were performed on single crystals of fully deuterated $κ-(ET)_2Cu[N(CN)_2]\,Br$, hereafter designated as $κ-(ET)_2Cu[N(CN)_2]\,Br$. This material phase separates into superconducting and antiferromagnetic regions, the degree of which depends strongly upon the cooling rate. Here, we show that the screening fraction $η_{sc}$ varies logarithmically with the cooling rate over nearly five decades. The average size of superconducting regions is estimated to vary from 5 to $40μm$, depending upon cooling rate, consistent with previous infrared microscopy measurements. In the region $T≲T_c/3$, the effective magnetic penetration depth exhibits power-law behavior $λ(T)-λ(0)~T^n$ with $n=1.6$, independent of the cooling rate. Changes in cooling rate and the consequent phase separation evidently do not introduce the kind of disorder that would alter the exponent $n$ in a $d$-wave superconductor. The exponent remains close to $n=$ 1.5, reported in single crystals of $κ-(ET)_2Cu[N(CN)_2]\,Br$ and $κ-(ET)_2Cu(NCS)_2$ [A. Carrington et al., Phys. Rev. Lett 83, 4172 (1999)]. The transition temperature fell linearly with $1-η_{sc}$. Measurements were also made on $κ-(ET)_2Cu[N(CN)_2]\,Cl$ of normal isotopic abundance in which a very small amount of superconducting phase $η_{sc}≈10^{-4}$ developed, presumably through the strain-induced sample mounting. This material showed a power-law exponent of $n=1.64$, independent of the cooling rate.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS2007

Nonthermal X-ray Spectral Flattening toward Low Energies in Early Impulsive Flares

The determination of the low-energy cutoff to nonthermal electron distributions is critical to the calculation of the nonthermal energy in solar flares. The most direct evidence for low-energy cutoffs is flattening of the power-law, nontherma1 X-ray spectra at low energies. However, because of the plasma preheating often seen in flares, the thermal emissions at low energies may hide such spectral flattening of the nonthermal component. We select a category of flares, which we call "early impulsive flares", in which the > 25 keV hard X-ray (HXR) flux increase is delayed by less than 30 s after the flux increase at lower energies. Thus, the plasma preheating in these flares is minimal, so the nonthermal spectrum can be determined to lower energies than in flares with significant preheating. Out of a sample of 33 early impulsive flares observed by the Ramaty High Energy Solar Spectroscopy Imager (RHESSI), 9 showed spectral flattening toward low energies. In these events, the break energy of the double power-law fit to the HXR spectra lies in the range of 10-50 keV, significantly lower than the value we have seen for other flares that do not show such early impulsive emissions. In particular, it correlates with the HXR flux. After correcting the spatially-integrated spectra for albedo from isotropically emitted X-rays and using RHESSI imaging spectroscopy to exclude the extended albedo halo, we find that albedo associated with isotropic or nearly isotropic electrons can only account for the spectral flattening in 3 flares near Sun center. The spectral flattening in the remaining 6 flares is found to be consistent with the existence of a low-energy cutoff in the electron spectrum, falling in the range of 15-50 keV, which also correlates with the HXR flux.

Holman, Gordon D.↗