arXiv subjects
Publications and source records attributed to M. Asplund.
We present the elemental abundances of HE1327-2326, the most iron-deficient star known, determined from a comprehensive analysis of spectra obtained with the Subaru Telescope High Dispersion Spectrograph.
The solar photospheric carbon abundance has been determined from [C I], C I, CH vibration-rotation, CH A-X electronic and C2 Swan electronic lines by means of a time-dependent, 3D, hydrodynamical model of the solar atmosphere. Departures from LTE have been considered for the C I lines. These turned out to be of increasing importance for stronger lines and are crucial to remove a trend in LTE abundances with the strengths of the lines. Very gratifying agreement is found among all the atomic and molecular abundance diagnostics in spite of their widely different line formation sensitivities. The mean of the solar carbon abundance based on the four primary abundance indicators ([C I], C I, CH vibration-rotation, C_2 Swan) is log C = 8.39 +/- 0.05, including our best estimate of possible systematic errors. Consistent results also come from the CH electronic lines, which we have relegated to a supporting role due to their sensitivity to the line broadening. The new 3D based solar C abundance is significantly lower than previously estimated in studies using 1D model atmospheres.
We present new observations of the center-to-limb variation of spectral lines in the quiet Sun. Our long-slit spectra are corrected for scattered light, which amounts to 4-8 % of the continuum intensity, by comparison with a Fourier transform spectrum of the disk center. We examine the effect of inelastic collisions with neutral hydrogen in NLTE line formation calculations of the oxygen infrared triplet, and the Na I 6160.8 A line. Adopting a classical one-dimensional theoretical model atmosphere, we find that the sodium transition, formed in higher layers, is much more effectively thermalized by hydrogen collisions than the high-excitation oxygen lines. This result appears as a simple consequence of the decrease of the ratio NH/Ne with depth in the solar photosphere. The center-to-limb variation of the selected lines is studied both under LTE and NLTE conditions. In the NLTE analysis, inelastic collisions with hydrogen atoms are considered with a simple approximation or neglected, in an attempt to test the validity of such approximation. For the sodium line studied, the best agreement between theory and observation happens when NLTE is considered and inelastic collisions with hydrogen are neglected in the rate equations. The analysis of the oxygen triplet benefits from a very detailed calculation using an LTE three-dimensional model atmosphere and NLTE line formation. The chi**2 statistics favors including hydrogen collisions with the approximation adopted, but the oxygen abundance derived in that case is significantly higher than the value derived from OH infrared transitions.
The solar photospheric oxygen abundance has been determined from [OI], OI, OH vibration-rotation and OH pure rotation lines by means of a realistic time-dependent, 3D, hydrodynamical model of the solar atmosphere. In the case of the OI lines, 3D non-LTE calculations have been performed, revealing significant departures from LTE as a result of photon losses in the lines. We derive a solar oxygen abundance of log O = 8.66 +/- 0.05. All oxygen diagnostics yield highly consistent abundances, in sharp contrast with the results of classical 1D model atmospheres. This low value is in good agreement with measurements of the local interstellar medium and nearby B stars. This low abundance is also supported by the excellent correspondence between lines of very different line formation sensitivities, and between the observed and predicted line shapes and center-to-limb variations. Together with the corresponding down-ward revisions of the solar carbon, nitrogen and neon abundances, the resulting significant decrease in solar metal mass fraction to Z = 0.0126 can, however, potentially spoil the impressive agreement between predicted and observed sound speed in the solar interior determined from helioseismology.
The possibility of unaccounted for opacity sources in the UV for late-type stars has often been invoked to explain discrepancies between predicted and observed flux distributions and spectral line strengths. Such missing UV-opacity could among other things have a significant impact on abundance determination for elements whose only relevant spectral features are accessible in this wavelength region, such as Be. Here, the study by Balachandran & Bell (1998) is re-visited in the light of a realistic 3D hydrodynamical solar model atmosphere and the recently significantly downward revised solar O abundance obtained with the same model atmosphere. The amount of missing UV-opacity, if any, is quantified by enforcing that the OH A-X electronic lines around 313 nm produce the same O abundance as the other available diagnostics: OH vibration-rotation and pure rotation lines in the IR, the forbidden [OI] 630.0 and 636.3 nm lines and high-excitation, permitted OI lines. This additional opacity is then applied for the synthesis of the BeII line at 313.0nm to derive a solar photospheric Be abundance in excellent agreement with the meteoritic value, thus re-enforcing the conclusions of Balachandran & Bell. The about 50% extra opacity over accounted for opacity sources can be well explained by recent calculations by the Iron Project for photo-ionization of FeI.
High resolution spectra of 34 halo population dwarf and subgiant stars have been obtained with VLT/UVES and used to derive sulphur abundances from the 8694.0, 8694.6 A and 9212.9, 9237.5 A SI lines. In addition, iron abundances have been determined from 19 FeII lines and zinc abundances from the 4722.2, 4810.5 ZnI lines. The abundances are based on a classical 1D, LTE model atmosphere analysis, but effects of 3D hydrodynamical modelling on the [S/Fe], [Zn/Fe] and [S/Zn] ratios are shown to be small. We find that most halo stars with metallicities in the range -3.2 < [Fe/H] < -0.8 have a near-constant [S/Fe] = +0.3; a least square fit to [S/Fe] vs. {Fe/H] shows a slope of only -0.04 +/- 0.01. Among halo stars with -1.2 < [Fe/H] < -0.8 the majority have [S/Fe] ~ +0.3, but two stars (previously shown to have low [alpha/Fe] ratios) have [S/Fe] ~ 0. For disk stars with [Fe/H] > -1, [S/Fe] decreases with increasing [Fe/H]. Hence, sulphur behaves like other typical alpha-capture elements, Mg, Si and Ca. Zinc, on the other hand, traces iron over three orders of magnitude in [Fe/H], although there is some evidence for a small systematic Zn overabundance ([Zn/Fe] ~ +0.1) among metal-poor disk stars and for halo stars with [Fe/H] < -2.0. Recent measurements of S and Zn in ten damped Ly-alpha systems (DLAs) with redshifts between 1.9 and 3.4 and zinc abundances in the range -2.1 < [Zn/H] < -0.15 show an offset relative to the [S/Zn] - [Zn/H] relation in Galactic stars. Possible reasons for this offset are discussed, including low and intermittent star formation rates in DLAs.
Rate coefficients for inelastic collisions between Li and H atoms covering all transitions between the asymptotic states Li(2s,2p,3s,3p,3d,4s,4p,4d,4f)+H(1s) and Li^+ +H^- are presented for the temperature range 2000-8000 K based on recent cross-section calculations. The data are of sufficient completeness for non-LTE modelling of the Li I 670.8 nm and 610.4 nm features in late-type stellar atmospheres. Non-LTE radiative transfer calculations in both 1D and 3D model atmospheres have been carried out for test cases of particular interest. Our detailed calculations show that the classical modified Drawin-formula for collisional excitation and de-excitation (Li*+H <-> Li*'+H) over-estimates the cross-sections by typically several orders of magnitude and consequently that these reactions are negligible for the line formation process. However, the charge transfer reactions collisional ion-pair production and mutual neutralization (Li*+H <-> Li^+ +H^-) are of importance in thermalizing Li. In particular, 3D non-LTE calculations of the Li I 670.8 nm line in metal-poor halo stars suggest that 1D non-LTE results over-estimate the Li abundance by up to about 0.1 dex, aggrevating the discrepancy between the observed Li abundances and the primordial Li abundance as inferred by the WMAP analysis of the cosmic microwave background.
The lithium abundances in metal-poor halo stars are of importance for cosmology, galaxy evolution and stellar structure. In an attempt to study possible systematic errors in the derived Li abundances, the line formation of LiI lines has been investigated by means of realistic 3D hydrodynamical model atmospheres of halo stars and 3D non-LTE radiative transfer calculations. These are the first detailed 3D non-LTE computations reported employing a multi-level atomic model showing that such problems are now computationally tractable. The detailed computations reveal that the LiI population has a strong influence from the radiation field rather than the local gas temperature, indicating that the low derived Li abundances found by Asplund et al. (1999) are an artifact of their assumption of LTE. Relative to 3D LTE, the detailed calculations show pronounced over-ionization. In terms of abundances the 3D non-LTE values are within 0.05 dex of the 1D non-LTE results for the particular cases of HD140283 and HD84937, which is a consequence of the dominance of the radiation in determining the population density of LiI. Although 3D non-LTE can be expected to give results rather close (+/-0.1 dex) to 1D non-LTE for this reason, there may be systematic trends with metallicity and effective temperature.
We report on the discovery of recently ionized hydrogen-deficient gas in the immediate circumstellar environment of the final helium shell flash star V4334 Sgr (Sakurai's object). On spectra obtained with FORS2 multi-object spectroscopy we have found spatially extended (about 2") emission from [N II], [O I], [O II] and very faint Halpha and [S II]. In the [N II] (ll6548,83) lines we have identified two components located at velocities -350 +/-50 and +200 +/-50 km/s, relative to V4334 Sgr itself. The full width of the [N II] l6583 feature at zero intensity corresponds to a velocity spread of about 1500 km/s. Based on the available data it is not possible to conclusively determine the mechanism of ionization. Both photo-ionization, from a rapidly evolving central star, and shock excitation, as the result of the collision of the fast ouflows with slower circumstellar matter, could account for the observed lines. The central star is still hidden behind strong dust absorption, since only a faint highly reddened continuum is apparent in the spectra. Theory states that it will become hotter and will retrace its post-asymptotic giant branch evolution towards the planetary nebula domain. Our detection of the ionized ejecta from the very late helium shell flash marks the beginning of a new phase in this star's amazingly rapid evolution.
From equivalent widths of the SI lines at 8694 A, Israelian & Rebolo (2001) and Takada-Hidai et al. (2002) have derived a surprisingly high sulphur-to-iron ratio ([S/Fe] = 0.5 to 0.7) in six halo stars with [Fe/H] ~ -2.0 suggesting perhaps that hypernovae made a significant contribution to the formation of elements in the early Galaxy. To investigate this problem we have used high-resolution spectra obtained with the ESO VLT/UVES spectrograph to determine the S/Fe ratio in 19 main-sequence and subgiant stars ranging in [Fe/H] from -3.2 to -0.7. The sulphur abundances are determined from SI lines at 8694 A and 9212 - 9237 A, and the iron abundances from about 20 FeII lines. S/Fe ratios as derived from 1D model atmospheres are presented and possible 3D effects are discussed. The initial results from our survey do not confirm the high values of [S/Fe] quoted above; instead we find that the ratio [S/Fe] remains constant at about 0.35 dex for metallicities -3 < [Fe/H] -1.
We have performed new determinations of sulphur and silicon abundances for a sample of 26 disk stars based on high-resolution, high signal-to-noise spectra. The results indicate a solar [S/Fe] for [Fe/H] >-0.3, below which [S/Fe] increases to ~0.25 dex at [Fe/H] =-1.0. We find that there is a good correlation between [S/H] and [Si/H], indicating the same nucleosynthetic origin of the two elements. It seems that the ratio of sulphur to silicon does not depend on metallicity for [Fe/H]> -1.0. The implications of these results on models for the nucleosynthesis of alpha-capture elements and the chemical evolution of the Galaxy are discussed.
A study of the O/Fe ratio in metal-poor main sequence and subgiant stars is presented using the [OI] 6300A line, the OI 7774A triplet, and a selection of weak FeII lines observed on high-resolution spectra acquired with the VLT UVES spectrograph. The [OI] line is detected in the spectra of 18 stars with -0.5 < [Fe/H] < -2.4, and the triplet is observed for 15 stars with [Fe/H] ranging from -1.0 to -2.7. The abundance analysis was made first using standard model atmospheres taking into account non-LTE effects on the triplet: the [OI] line and the triplet give consistent results with [O/Fe] increasing quasi-linearly with decreasing [Fe/H] reaching [O/Fe] ~ +0.7 at [Fe/H] = -2.5. When hydrodynamical model atmospheres representing stellar granulation in dwarf and subgiant stars replace standard models, the [O/Fe] from the [OI] and FeII lines is decreased by an amount which increases with decreasing [Fe/H]. The [O/Fe] vs [Fe/H] relation remains quasi-linear extending to [O/Fe] ~ +0.5 at [Fe/H] = -2.5, but with a tendency of a plateau with [O/Fe] ~ +0.3 for -2.0 < [Fe/H] < -1.0, and a hint of cosmic scatter in [O/Fe] at [Fe/H] ~ -1.0. Use of the hydrodynamical models disturbs the broad agreement between the oxygen abundances from the [OI], OI, and OH lines, but 3D non-LTE effects may serve to erase these differences.
The formation of the UV OH spectral lines has been investigated for a range of stellar parameters in the light of 3D hydrodynamical model atmospheres. The low atmospheric temperatures encountered at low metallicities compared with the radiative equilibrium values enforced in classical 1D hydrostatic model atmospheres have a profound impact on the OH line strengths. As a consequence, the derived O abundances using 3D models are found to be systematically lower by more than 0.6 dex at [Fe/H]=-3.0 compared with previous 1D analyses, casting doubts on the recent claims for a monotonic increase in [O/Fe] towards lower metallicities. In fact, taken at face value the resulting 3D LTE trend is in rough agreement with the conventional [O/Fe] plateau. Caution must, however, be exercised in view of the remaining assumptions in the 3D calculations. We have verified that the stellar parameters remain essentially unchanged with 3D model atmospheres provided that the infrared flux method, Hipparcos parallaxes and FeII lines are utilised, leaving the 3D O abundances from OH lines largely intact. Greater concern stems from possible departures from LTE in both the line formation and the molecular equilibrium, which, if present, would increase the derived O abundances again. Non-LTE line formation calculations with 1D model atmospheres suggest no significant steepening of the [O/Fe] trend even if the abundance corrections amount to about 0.2 dex for all investigated stellar parameters. We note, however, that the 3D case may not necessarily be as metallicity-independent. The apparent lack of laboratory or theoretical rate coefficients at the relevant temperatures for the involved molecular reactions unfortunately prevents a quantitative discussion on the possible effects of non-equilibrium chemistry.
The formation of [O I], O I and OH lines in metal-poor stars has been studied by means of 3D hydrodynamical model atmospheres. For O I detailed 3D non-LTE calculations have been performed. While the influence of 3D model atmospheres is minor for [O I] and O I lines, the very low temperatures encountered at low metallicities have a drastic impact on the OH lines. As a result, the derived O abundances are found to be systematically overestimated in 1D analyses, casting doubts on the recent claims for a monotonic increase in [O/Fe] towards lower metallicities.
We report on a new spectroscopic analysis of the very metal deficient star G 64-12 ([Fe/H]=-3.3), aimed at determining, for the first time, its Be content. The spectra were observed during the Science Verification of UVES, the ESO VLT Ultraviolet and Visible Echelle Spectrograph. The high resolution (~48,000) and high S/N (~130 per pixel) achieved at the wavelengths of the BeII resonance doublet allowed an accurate determination of its abundance: log N(Be/H) = -13.10 +/- 0.15 dex. The Be abundance is significantly higher than expected from previous measurements of Be in stars of similar metallicity (3D and NLTE corrections acting to make a slightly higher value than an LTE analysis). When compared to iron, the high [Be/Fe] ratio thus found may suggest a flattening in the beryllium evolutionary trend at the lowest metallicity end or the presence of dispersion at early epochs of galactic evolution.
The computationally demanding nature of radiative-hydrodynamical simulations of stellar surface convection warrants an investigation of the sensitivity of the convective structure and spectral synthesis to the numerical resolution and dimension of the simulations, which is presented here. With too coarse a resolution the predicted spectral lines tend to be too narrow, reflecting insufficient Doppler broadening from the convective motions, while at the currently highest affordable resolution the line shapes have converged essentially perfectly to the observed profiles. Similar conclusions are drawn from the line asymmetries and shifts. In terms of abundances, weak FeI and FeII lines show a very small dependence (~0.02 dex) while for intermediate strong lines with significant non-thermal broadening the sensitivity increases (~0.10 dex). Problems arise when using 2D convection simulations to describe an inherent 3D phenomenon, which translates to inaccurate atmospheric velocity fields and temperature and pressure structures. In 2D the theoretical line profiles tend to be too shallow and broad compared with the 3D calculations and observations, in particular for intermediate strong lines. In terms of abundances, the 2D results are systematically about 0.1 dex lower than for the 3D case for FeI lines. Furthermore, the predicted line asymmetries and shifts are much inferior in 2D. Given these shortcomings and computing time considerations it is better to use 3D simulations of even modest resolution than high-resolution 2D simulations.
Realistic ab-initio 3D, radiative-hydrodynamical convection simulations of the solar granulation have been applied to FeI and FeII line formation. In contrast to classical analyses based on hydrostatic 1D model atmospheres the procedure contains no adjustable free parameters but the treatment of the numerical viscosity in the construction of the 3D, time-dependent, inhomogeneous model atmosphere and the elemental abundance in the 3D spectral synthesis. However, the numerical viscosity is introduced purely for numerical stability purposes and is determined from standard hydrodynamical test cases with no adjustments allowed to improve the agreement with the observational constraints from the solar granulation. The non-thermal line broadening is mainly provided by the Doppler shifts arising from the convective flows in the solar photosphere and the solar oscillations. The almost perfect agreement between the predicted temporally and spatially averaged line profiles for weak Fe lines with the observed profiles and the absence of trends in derived abundances with line strengths, seem to imply that the micro- and macroturbulence concepts are obsolete in these 3D analyses. Furthermore, the theoretical line asymmetries and shifts show a very satisfactory agreement with observations with an accuracy of typically 50-100 m/s on an absolute velocity scale. The remaining minor discrepancies point to how the convection simulations can be refined further.
The solar photospheric Fe abundance has been determined using realistic ab initio 3D, time-dependent, hydrodynamical model atmospheres. The study is based on the excellent agreement between the predicted and observed line profiles directly rather than equivalent width, since the intrinsic Doppler broadening from the convective motions and oscillations provide the necessary non-thermal broadening. Thus, three of the four hotly debated parameters (equivalent widths, microturbulence and damping enhancement factors) in the center of the recent solar Fe abundance dispute regarding FeI lines no longer enter the analysis, leaving the transition probabilities as the main uncertainty. Both FeI (using the samples of lines of both the Oxford and Kiel studies) and FeII lines have been investigated, which give consistent results: log FeI = 7.44 +- 0.05 and log FeII = 7.45 +- 0.10. Also the wings of strong FeI lines return consistent abundances, log FeII = 7.42 +- 0.03, but due to the uncertainties inherent in analyses of strong lines we give this determination lower weight than the results from weak and intermediate strong lines. In view of the recent slight downward revision of the meteoritic Fe abundance log Fe = 7.46 +- 0.01, the agreement between the meteoritic and photospheric values is very good, thus appearingly settling the debate over the photospheric Fe abundance from FeI lines.