arXiv subjects

Knox S. Long

Publications and source records attributed to Knox S. Long.

At least 109 records · Page 6Linked to original sources

arXiv2012

X-ray Proper Motions and Shock Speeds along the Northwest Rim of SN 1006

We report the results of an X-ray proper motion measurement for the NW rim of SN1006, carried out by comparing Chandra observations from 2001 and 2012. The NW limb has predominantly thermal X-ray emission, and it is the only location in SN1006 with significant optical emission: a thin, Balmer-dominated filament. For most of the NW rim, the proper motion is about 0.30 arcsec/yr, essentially the same as has been measured from the H-alpha filament. Isolated regions of the NW limb are dominated by nonthermal emission, and here the proper motion is much higher, 0.49 arcsec/yr, close to the value measured in X-rays along the much brighter NE limb, where the X-rays are overwhelmingly nonthermal. At the 2.2 kpc distance to SN1006, the proper motions imply shock velocities of about 3000 km/s and 5000 km/s in the thermal and nonthermal regions, respectively. A lower velocity behind the H-alpha filament is consistent with the picture that SN1006 is encountering denser gas in the NW, as is also suggested by its overall morphology. In the thermally-dominated portion of the X-ray shell, we also see an offset in the radial profiles at different energies; the 0.5-0.6 keV peak dominated by O VII is closer to the shock front than that of the 0.8-3 keV emission--due to the longer times for heavier elements to reach ionization states where they produce strong X-ray emission.

astro-ph.GA↗

arXiv2012

Recovery of the Historical SN1957D in X-rays with Chandra

SN1957D, located in one of the spiral arms of M83, is one of the small number of extragalactic supernovae that has remained detectable at radio and optical wavelengths during the decades after its explosion. Here we report the first detection of SN1957D in X-rays, as part of a 729 ks observation of M83 with \chandra. The X-ray luminosity (0.3 - 8 keV) is 1.7 (+2.4,-0.3) 10**37 ergs/s. The spectrum is hard and highly self-absorbed compared to most sources in M83 and to other young supernova remnants, suggesting that the system is dominated at X-ray wavelengths by an energetic pulsar and its pulsar wind nebula. The high column density may be due to absorption within the SN ejecta. HST WFC3 images resolve the supernova remnant from the surrounding emission and the local star field. Photometry of stars around SN1957D, using WFC3 images, indicates an age of less than 10**7 years and a main sequence turnoff mass more than 17 solar masses. New spectra obtained with Gemini-South show that the optical spectrum continues to be dominated by broad [O III] emission lines, the signature of fast-moving SN ejecta. The width of the broad lines has remained about 2700 km/s (FWHM). The [O III] flux dropped precipitously between 1989 and 1991, but continued monitoring shows the flux has been almost constant since. In contrast, radio observations over the period 1990-2011 show a decline rate inf the flux proportional to t**-4, far steeper than the rate observed earlier, suggesting that the primary shock has overrun the edge of a pre-SN wind.

astro-ph.HE↗

arXiv2012

Dust in a Type Ia Supernova Progenitor: Spitzer Spectroscopy of Kepler's Supernova Remnant

Characterization of the relatively poorly-understood progenitor systems of Type Ia supernovae is of great importance in astrophysics, particularly given the important cosmological role that these supernovae play. Kepler's Supernova Remnant, the result of a Type Ia supernova, shows evidence for an interaction with a dense circumstellar medium (CSM), suggesting a single-degenerate progenitor system. We present 7.5-38 $μ$m infrared (IR) spectra of the remnant, obtained with the {\it Spitzer Space Telescope}, dominated by emission from warm dust. Broad spectral features at 10 and 18 $μ$m, consistent with various silicate particles, are seen throughout. These silicates were likely formed in the stellar outflow from the progenitor system during the AGB stage of evolution, and imply an oxygen-rich chemistry. In addition to silicate dust, a second component, possibly carbonaceous dust, is necessary to account for the short-wavelength IRS and IRAC data. This could imply a mixed chemistry in the atmosphere of the progenitor system. However, non-spherical metallic iron inclusions within silicate grains provide an alternative solution. Models of collisionally-heated dust emission from fast shocks ($>$ 1000 km s$^{-1}$) propagating into the CSM can reproduce the majority of the emission associated with non-radiative filaments, where dust temperatures are $\sim 80-100$ K, but fail to account for the highest temperatures detected, in excess of 150 K. We find that slower shocks (a few hundred km s$^{-1}$) into moderate density material ($n_{0} \sim 50-250$ cm$^{-3}$) are the only viable source of heating for this hottest dust. We confirm the finding of an overall density gradient, with densities in the north being an order of magnitude greater than those in the south.

astro-ph.GA↗

arXiv2012

The Birth of an Ultra-Luminous X-ray Source in M83

A previously undetected X-ray source (L_X<10**36 erg/s) in the strongly star-forming galaxy M83 entered an ultraluminous state between August 2009 and December 2010. It was first seen with Chandra on 23 December 2010 at L_X ~ 4 10**39 ergs/s, and has remained ultraluminous through our most recent observations in December 2011, with typical flux variation of a factor of two. The spectrum is well fitted by a combination of absorbed power-law and disk black-body models. While the relative contributions of the models varies with time, we have seen no evidence for a canonical state transition. The luminosity and spectral properties are consistent with accretion powered by a black hole with M_BH ~ 40-100 solar masses. In July 2011 we found a luminous, blue optical counterpart which had not been seen in deep HST observations obtained in August 2009. These optical observations suggest that the donor star is a low-mass star undergoing Roche-lobe overflow, and that the blue optical emission seen during the outburst is coming from an irradiated accretion disk. This source shows that ultraluminous X-ray sources (ULXs) with low-mass companions are an important component of the ULX population in star-forming galaxies, and provides further evidence that the blue optical counterparts of some ULXs need not indicate a young, high-mass companion, but rather that they may indicate X-ray reprocessing.

astro-ph.HE↗

arXiv2012

Spitzer Imaging and Spectral Mapping of the Oxygen-Rich Supernova Remnant G292.0+1.8

We present mid-infrared continuum and emission line images of the Galactic oxygen-rich supernova remnant (SNR) G292.0+1.8, acquired using the MIPS and IRS instruments on the Spitzer Space Telescope. The MIPS 24 micron and 70 micron images of G292.0+1.8 are dominated by continuum emission from a network of filaments encircling the SNR. The morphology of the SNR, as seen in the mid-infrared, resembles that seen in X-rays with the Chandra X-ray Observatory. Most of the mid-infrared emission in the MIPS images is produced by circumstellar dust heated in the non-radiative shocks around G292.0+1.8, confirming the results of earlier mid-IR observations with AKARI. In addition to emission from hot dust, we have also mapped atomic line emission between 14 micron and 36 micron using IRS spectral maps. The line emission is primarily associated with the bright oxygen-rich optical knots, but is also detected from fast-moving knots of ejecta. We confirm our earlier detection of 15-25 micron emission characteristic of magnesium silicate dust in spectra of the radiatively shocked ejecta. We do not detect silicon line emission from any of the radiatively shocked ejecta in the southeast of the SNR, possibly because that the reverse shock has not yet penetrated most of the Si-rich ejecta in that region. This may indicate that G292.0+1.8 is less evolved in the southeast than the rest of the SNR, and may be further evidence in favor of an asymmetric SN explosion as proposed in recent X-ray studies of G292.0+1.8.

astro-ph.GA↗

arXiv2011

Modelling the spectra of (BAL)QSOs

We have embarked upon a project to model the UV spectra of BALQSOs using a Monte Carlo radiative transfer code previously validated through modelling of the winds of cataclysmic variable stars (e.g. Noebauer et al. 2010). We intend to use the simulations to investigate the plausibility of geometric unification (e.g. Elvis 2000) of the different classes of QSO. Here we introduce the code, and present some initial results. These demonstrate that for reasonable geometries and mass loss rates we are able to produce synthetic spectra which reproduce the important features of observed BALQSO spectra.

astro-ph.CO↗

arXiv2011

Time Evolution of the Reverse Shock in SN 1006

The Schweizer-Middleditch star, located behind the SN 1006 remnant and near its center in projection, provides the opportunity to study cold, expanding ejecta within the SN 1006 shell through UV absorption. Especially notable is an extremely sharp red edge to the Si II 1260 Angstrom feature, which stems from the fastest moving ejecta on the far side of the SN 1006 shell--material that is just encountering the reverse shock. Comparing HST far-UV spectra obtained with COS in 2010 and with STIS in 1999, we have measured the change in this feature over the intervening 10.5-year baseline. We find that the sharp red edge of the Si II feature has shifted blueward by 0.19 +/- 0.05 Angstroms, which means that the material hitting the reverse shock in 2010 was moving slower by 44 +/- 11 km/s than the material that was hitting it in 1999, a change corresponding to - 4.2 +/- 1.0 km/s/yr. This is the first observational confirmation of a long-predicted dynamic effect for a reverse shock: that the shock will work its way inward through expanding supernova ejecta and encounter ever slower material as it proceeds. We also find that the column density of shocked Si II (material that has passed through the reverse shock) has decreased by 7 +/- 2% over the ten-year period. The decrease could indicate that in this direction the reverse shock has been ploughing through a dense clump of Si,leading to pressure and density transients.

astro-ph.GA↗

arXiv2011

RCW 86: A Type Ia Supernova in a Wind-Blown Bubble

We report results from a multi-wavelength analysis of the Galactic SNR RCW 86, the proposed remnant of the supernova of 185 A.D. We report new infrared observations from {\it Spitzer} and {\it WISE}, where the entire shell is detected at 24 and 22 $μ$m. We fit the infrared flux ratios with models of collisionally heated ambient dust, finding post-shock gas densities in the non-radiative shocks of 2.4 and 2.0 cm$^{-3}$ in the SW and NW portions of the remnant, respectively. The Balmer-dominated shocks around the periphery of the shell, large amount of iron in the X-ray emitting ejecta, and lack of a compact remnant support a Type Ia origin for this remnant. From hydrodynamic simulations, the observed characteristics of RCW 86 are successfully reproduced by an off-center explosion in a low-density cavity carved by the progenitor system. This would make RCW 86 the first known case of a Type Ia supernova in a wind-blown bubble. The fast shocks ($> 3000$ km s$^{-1}$) observed in the NE are propagating in the low-density bubble, where the shock is just beginning to encounter the shell, while the slower shocks elsewhere have already encountered the bubble wall. The diffuse nature of the synchrotron emission in the SW and NW is due to electrons that were accelerated early in the lifetime of the remnant, when the shock was still in the bubble. Electrons in a bubble could produce gamma-rays by inverse-Compton scattering. The wind-blown bubble scenario requires a single-degenerate progenitor, which should leave behind a companion star.

astro-ph.GA↗

arXiv2011

The cataclysmic variable SDSS J1507+52: An eclipsing period bouncer in the Galactic halo

SDSS J1507+52 is an eclipsing cataclysmic variable consisting of a cool, non-radially pulsating white dwarf and an unusually small sub-stellar secondary. The system has a high space velocity and a very short orbital period of about 67 minutes, well below the usual minimum period for CVs. To explain the existence of this peculiar system, two theories have been proposed. One suggests that SDSS J1507+52 was formed from a detached white-dwarf/brown-dwarf binary. The other theory proposes that the system is a member of the Galactic halo-population. Here, we present ultraviolet spectroscopy of SDSS J1507+52 obtained with the Hubble Space Telescope with the aim of distinguishing between these two theories. The UV flux of the system is dominated by emission from the accreting white dwarf. Fits to model stellar atmospheres yield physical parameter estimates of T(eff) = 14200 \pm 500 K, log(g)=8.2 \pm 0.3, vsin(i)=180 \pm 20 kms-1 and [Fe/H]=-1.2 \pm 0.2. These fits suggest a distance towards SDSS J1507+52 of d = 250 \pm 50 pc. The quoted uncertainties include systematic errors associated with the adopted fitting windows and interstellar reddening. Assuming that there is no contribution to the UV flux from a hot, optically thick boundary layer, we find a T(eff) much higher than previously estimated from eclipse analysis. The strongly sub-solar metallicity we infer for SDSS J1507+52 is consistent with that of halo stars at the same space velocity. We therefore conclude that SDSS J1507+52 is a member of the Galactic halo.

astro-ph.SR↗

arXiv2011

Dusty Blastwaves of Two Young LMC Supernova Remnants: Constraints on Postshock Compression

We present results from mid-IR spectroscopic observations of two young supernova remnants (SNRs) in the Large Magellanic Cloud (LMC) done with the {\it Spitzer Space Telescope}. We imaged SNRs B0509-67.5 and B0519-69.0 with {\it Spitzer} in 2005, and follow-up spectroscopy presented here confirms the presence of warm, shock heated dust, with no lines present in the spectrum. We use model fits to {\it Spitzer} IRS data to estimate the density of the postshock gas. Both remnants show asymmetries in the infrared images, and we interpret bright spots as places where the forward shock is running into material that is several times denser than elsewhere. The densities we infer for these objects depend on the grain omposition assumed, and we explore the effects of differing grain porosity on the model fits. We also analyze archival {\it XMM-Newton} RGS spectroscopic data, where both SNRs show strong lines of both Fe and Si, coming from ejecta, as well as strong O lines, which may come from ejecta or shocked ambient medium. We use model fits to IRS spectra to predict X-ray O line strengths for various grain models and values of the shock compression ratio. For 0509-67.5, we find that compact (solid) grain models require nearly all O lines in X-ray spectra to originate in reverse-shocked ejecta. Porous dust grains would lower the strength of ejecta lines relative to those arising in the shocked ambient medium. In 0519-69.0, we find significant evidence for a higher than standard compression ratio of 12, implying efficient cosmic-ray acceleration by the blast wave. A compact grain model is favored over porous grain models. We find that the dust-to-gas mass ratio of the ambient medium is significantly lower than what is expected in the ISM.

astro-ph.SR↗

arXiv2010

Steady X-Ray Synchrotron Emission in the Northeastern Limb of SN 1006

We investigate time variations and detailed spatial structures of X-ray synchrotron emission in the northeastern limb of SN 1006, using two Chandra observations taken in 2000 and 2008. We extract spectra from a number of small (about 10") regions. After taking account of proper motion and isolating the synchrotron from the thermal emission, we study time variations in the synchrotron emission in the small regions. We find that there are no regions showing strong flux variations. Our analysis shows an apparent flux decline in the overall synchrotron flux of about 4% at high energies, but we suspect that this is mostly a calibration effect, and that flux is actually constant to about 1%. This is much less than the variation found in other remnants where it was used to infer magnetic-field strengths up to 1 mG. We attribute the lack of variability to the smoothness of the synchrotron morphology, in contrast to the small-scale knots found to be variable in other remnants. The smoothness is to be expected for a Type Ia remnant encountering uniform material. Finally we find a spatial correlation between the flux and the cut-off frequency in synchrotron emission. The simplest interpretation is that the cut-off frequency depends on the magnetic-field strength. This would require that the maximum energy of accelerated electrons is not limited by synchrotron losses, but by some other effect. Alternatively, the rate of particle injection and acceleration may vary due to some effect not yet accounted for, such as a dependence on shock obliquity.

astro-ph.HE↗

arXiv2010

The Geometry and Ionization Structure of the Wind in the Eclipsing Nova-like Variables RW Tri and UX UMa

The UV spectra of nova-like variables are dominated by emission from the accretion disk, modified by scattering in a wind emanating from the disk. Here we model the spectra of RW Tri and UX UMa, the only two eclipsing nova-likes which have been observed with the Hubble Space Telescope in the far-ultraviolet, in an attempt to constrain the geometry and the ionization structure of their winds. Using our Monte Carlo radiative transfer code we computed spectra for simply-parameterized axisymmetric biconical outflow models and were able to find plausible models for both systems. These reproduce the primary UV resonance lines - N V, Si IV, and C IV - in the observed spectra in and out of eclipse. The distribution of these ions in the wind models is similar in both cases as is the extent of the primary scattering regions in which these lines are formed. The inferred mass loss rates are 6% to 8% of the mass accretion rates for the systems. We discuss the implication of our point models for our understanding of accretion disk winds in cataclysmic variables.

astro-ph.SR↗

arXiv2010

Large amplitude variability from the persistent ultracompact X-ray binary in NGC 1851

Using archival RXTE data, we show that the ultracompact X-ray binary in NGC 1851 exhibits large amplitude X-ray flux varations of more than a factor of 10 on timescales of days to weeks and undergoes sustained periods of months where the time-averaged luminosty varies by factors of two. Variations of this magnitude and timescale have not been reported previously in other ultracompact X-ray binaries. Mass transfer in ultracompact binaries is thought to be driven by gravitational radiation and the predicted transfer rates are so high that the disks of ultracompact binaries with orbits as short as that of this object should not be susceptible to ionization instabilities. Therefore the variability characteristics we observe were unexpected, and need to be understood. We briefly discuss a few alternatives for producing the observed variations in light of the fact that the viscous timescale of the disk is of order a week, comparable to the shorter time scale variation that is observed but much less than the longer term variation. We also discuss the implications for interpretation of observations of extragalactic binaries if the type of variability seen in the source in NGC 1851 is typical.

astro-ph.HE↗

arXiv2010

The Chandra ACIS Survey of M33: X-ray, Optical and Radio Properties of the Supernova Remnants

M33 contains a large number of emission nebulae identified as supernova remnants (SNRs) based on the high [S II]:Ha ratios characteristic of shocked gas. Using Chandra data from the ChASeM33 survey with a 0.35-2 keV sensitivity of about 2 x 10**34 ergs/s, we have detected 82 of 137 SNR candidates, yielding confirmation of (or at least strongly support for) their SNR identifications. This provides the largest sample of remnants detected at optical and X-ray wavelengths in any galaxy, including the Milky Way. A spectral analysis of the seven X-ray brightest SNRs reveals that two, G98-31 and G98-35, have spectra that appear to indicate enrichment by ejecta from core-collapse supernova explosions. In general, the X-ray detected SNRs have soft X-ray spectra compared to the vast majority of sources detected along the line of sight to M33. It is unlikely that there are any other undiscovered thermally dominated X-ray SNRs with luminosities in excess of about 4 x 10**35 ergs/s in the portions of M33 covered by the ChASeM33 survey. We have used a combination of new and archival optical and radio observations to attempt to better understand why some objects are detected as X-ray sources and others are not. We have also developed a morphological classification scheme for the optically-identified SNRs, and discuss the efficacy of this scheme as a predictor of X-ray detectability. Finally, we have compared the SNRs found in M33 to those that have been observed in the Galaxy and the Magellanic Clouds. There are no close analogs of Cas A, Kepler's SNR, Tycho's SNR or the Crab Nebula in the regions of M33 surveyed, but we have found an X-ray source with a power law spectrum coincident with a small-diameter radio source that may be the first pulsar-wind nebula recognized in M33.

astro-ph.GA↗

arXiv2010

A Far-UV Survey of M 80: X-ray Source Counterparts, Strange Blue Stragglers and the Recovery of Nova T Sco

Using the ACS on HST, we have surveyed the FUV and NUV populations in the core region of M80. The CMD reveals large numbers of blue and extreme horizontal branch stars and blue stragglers, as well as approx. 60 objects lying in the region of the CMD where accreting and detached white dwarf binaries are expected. Overall, the blue straggler stars are the most centrally concentrated population, with their radial distribution suggesting a typical blue straggler mass of about 1.2 Msun. However, counterintuitively, the faint blue stragglers are significantly more centrally concentrated than the bright ones and a Kolmogorov-Smirnov test suggest only a 3.5% probability that both faint and bright blue stragglers are drawn from the same distribution. This may suggest that (some) blue stragglers get a kick during their formation. We have also been able to identify the majority of the known X-ray sources in the core with FUV bright stars. One of these FUV sources is a likely dwarf nova that was in eruption at the time of the FUV observations. This object is located at a position consistent with Nova 1860 AD, or T Scorpii. Based on its position, X-ray and UV characteristics, this system is almost certainly the source of the nova explosion. The radial distribution of the X-ray sources and of the cataclysmic variable candidates in our sample suggest masses > 1 Msun.

astro-ph.SR↗

arXiv2010

Supernova Remnants and the Interstellar Medium of M83: Imaging & Photometry with WFC3 on HST

We present Wide Field Camera 3 images taken with the Hubble Space Telescope within a single field in the southern grand design star-forming galaxy M83. Based on their size, morphology and photometry in continuum-subtracted H$α$, [\SII], H$β$, [\OIII] and [\OII] filters, we have identified 60 supernova remnant candidates, as well as a handful of young ejecta-dominated candidates. A catalog of these remnants, their sizes and, where possible their H$α$ fluxes are given. Radiative ages and pre-shock densities are derived from those SNR which have good photometry. The ages lie in the range $2.62 < log(τ_{\rm rad}/{\rm yr}) < 5.0$, and the pre-shock densities at the blast wave range over $0.56 < n_0/{\rm cm^{-3}} < 1680$. Two populations of SNR have been discovered. These divide into a nuclear and spiral arm group and an inter-arm population. We infer an arm to inter-arm density contrast of 4. The surface flux in diffuse X-rays is correlated with the inferred pre-shock density, indicating that the warm interstellar medium is pressurised by the hot X-ray plasma. We also find that the interstellar medium in the nuclear region of M83 is characterized by a very high porosity and pressure and infer a SNR rate of one per 70-150 yr for the nuclear ($R<300 $pc) region. On the basis of the number of SNR detected and their radiative ages, we infer that the lower mass of Type II SNe in M83 is $M_{\rm min} = 16^{+7}_ {-5}$ M$_{\odot}$. Finally we give evidence for the likely detection of the remnant of the historical supernova, SN1968L.

astro-ph.CO↗

arXiv2009

The Effect of a Superoutburst on the White Dwarf and Disk of VW Hydri as observed with FUSE

We have used FUSE to obtain thirteen observations of the dwarf nova VW Hyi covering the period from the end of a superoutburst through the following normal outburst of the system. Here, we present the quiescent spectra. They contain at least three components. The dominant component is the white dwarf (WD), which cools following the superoutburst. The amount of cooling depends somewhat on the WD models used. For log g of 8.0, the temperature drops from 24,000K just after the outburst to 20,000 K just before the normal outburst. For this model, and a distance of 65 pc, the WD radius is ~8 x 10**8 cm and v sin(i) is ~420 km/s. The fact that the derived radius is smaller than expected for a WD with log g=8 suggests a higher gravity WD or that VW Hyi is somewhat further than 65 pc. Either is possible given the current distance uncertainty of +-20 pc. Earlier suggestions that the WD photosphere shows evidence of CNO processed material are confirmed, but our analysis also shows that issues remain in terms of analyzing the spectra of WDs in such unusual physical situations. The second component is relatively featureless and shows modulation on the orbital (and just after outburst, the superhump) period. It is likely associated with the hot spot where material from the secondary encounters the disk, rather than the boundary layer region between the inner disk and WD. The second component fades ~10 days after the superoutburst. There is also a third component, clearly visible in broad emission lines of C III 977, N III 991, Lyman Beta and O VI 1032,1038, which appears to be accompanied by a flat continuum. The strength of the emission lines, which are almost surely associated with the accretion disk, appear relatively constant throughout the observations.

astro-ph.GA↗

arXiv2009

Spitzer Spectroscopy of the Galactic Supernova Remnant G292.0+1.8: Structure and Composition of the Oxygen-Rich Ejecta

We present mid-infrared (5-40 micron) spectra of shocked ejecta in the Galactic oxygen-rich supernova remnant G292.0+1.8, acquired with the IRS spectrograph on board the Spitzer Space Telescope. The observations targeted two positions within the brightest oxygen-rich feature in G292.0+1.8. Emission lines of [Ne II] 12.8, [Ne III] 15.5, 36.0, [Ne V] 24.3 and [O IV] 25.9 are detected from the shocked ejecta. No discernible mid-IR emission from heavier species such as Mg, Si, S, Ar or Fe is detected in G292.0+1.8. We also detect a broad emission bump between 15 and 28 microns in spectra of the radiatively shocked O-rich ejecta in G292.0+1.8. We suggest that this feature arises from either shock-heated Mg2SiO4 (forsterite) dust in the radiatively shocked O-rich ejecta, or collisional excitation of PAHs in the blast wave of the SNR. If the former interpretation is correct, this would be the first mid-IR detection of ejecta dust in G292.0+1.8. A featureless dust continuum is also detected from non-radiative shocks in the circumstellar medium around G292.0+1.8. The mid-IR continuum from these structures is well described by a two-component dust model. The temperature of the hot dust component (mass ~ 0.002 Solar masses) is ~ 115 K, while that of the cold component (> = 3 Solar masses) is ~ 35 K. We attribute the hot component to collisionally heated dust in the circumstellar shocks in G292.0+1.8, and attribute the cold component to dust heated by the hard FUV radiation from the circumstellar shocks. Our models yield mid-IR line strengths consistent with M(O)/M(Ne) ~ 3, M(O)/M(Si) >~ 61 and M(O)/M(S) ~ 50. These ratios are difficult to reproduce with standard nucleosynthesis models of well-mixed SN ejecta (abridged).

astro-ph.GA↗