Engineering topics
Publications and source records attributed to Skinner, C. J..
Models are presented of four Vega-like stars: main-sequence stars with infrared emission from circumstellar dust. The dusty environments of the four stars are rather diverse, as shown by their spectral energy distributions. Good fits to the observations were obtained for all four stars.
Silicon carbide (SiC) is known to form in circumstellar shells around carbon stars. SiC can come in two basic types - hexagonal alpha-SiC or cubic beta-SiC. Laboratory studies have shown that both types of SiC exhibit an emission feature in the 11-11.5 micron region, the size and shape of the feature varying with type, size and shape of the SiC grains. Such a feature can be seen in the spectra of carbon stars. Silicon carbide grains have also been found in meteorites. The aim of the current work is to identity the type(s) of SiC found in circumstellar shells and how they might relate to meteoritic SiC samples. We have used the CGS3 spectrometer at the 3.8 m UKIRT to obtain 7.5-13.5 micron spectra of 31 definite or proposed carbon stars. After flux-calibration, each spectrum was fitted using a chi(exp 2)-minimisation routine equipped with the published laboratory optical constants of six different samples of small SiC particles, together with the ability to fit the underlying continuum using a range of grain emissivity laws. It was found that the majority of observed SiC emission features could only be fitted by alpha-SiC grains. The lack of beta-SiC is surprising, as this is the form most commonly found in meteorites. Included in the sample were four sources, all of which have been proposed to be carbon stars, that appear to show the SiC feature in absorption.
We have observed the extreme OH/IR star, OH 26.5+0.6, in the infrared dust continuum and in the sub- millimeter rotational lines of CO. Mid-infrared images reveal the compact nature of the circumstellar shell (less than 0.5 sec). A deep 9.7 microns absorption feature and an absorption at 18 microns show that the dust mass-loss rate is very high. However, the low antenna temperatures of CO J = 1-0 and 2-1 lines suggest that the outer part of the circumstellar shell is much more tenuous. In order to resolve this discrepancy, we have observed the J = 3-2 and 4-3 CO rotational transitions. We have developed a model for the circumstellar shell for OH 26.5 + 0.6 which is consistent with the infrared and submillimeter observations. The dust and gas data are well fitted by a two-shell model, consisting of a dense shell surrounded by a more tenuous shell. The former we identify with the superwind (M = 5.5 x 10(exp -4) solar mass/ yr), and the latter we identify with mass loss on the asymptotic giant branch (AGB) (M = 10(exp -6) solar mass/ yr). The transition between the two mass-loss phases is shown to be rather abrupt ((Delta)t less than 150 yr). Depending on the mass of the progenitor, this superwind phase may be the last thermal pulse (for M(sub *) less than 1.5 solar mass), or the first of a series of the superwind phases (for up to 8 solar mass), punctuated by a period of low mass-loss rates, before the star evolves off the AGB.
A significant fraction of a star's initial mass is lost while it is on the Asymptotic Giant Branch (AGB). Mass loss rates range from 10(exp -7) solar mass/yr for early AGB stars to a few 10(exp -4) solar mass/yr for stars at the tip of the AGB. Dust grains condense from the outflow as the gas expands and form a dust shell around the central star. A superwind (approximately 10(exp -4) to 10(exp -3) solar mass/yr) is thought to terminate the AGB phase. In the post-AGB phase, the star evolves to a higher effective temperature, the mass loss decreases (approximately 10(exp -8) solar mass/yr), but the wind velocity increases (approximately 1000 km/s). During this evolution, dust and gas are exposed to an increasingly harsher radiation field and when T(sub eff) reaches about 30,000 K, the nebula is ionized and becomes a planetary nebula (PN). Photons from the central star can create a photodissociation region (PDR) in the expanding superwind. Gas can be heated through the photoelectric effect working on small grains and polycyclic aromatic hydrocarbons (PAH's). This gas can cool via the atomic fine structure lines of O I (63 microns and 145 microns) and C II (158 microns), as well as the rotational lines of CO. In the post-AGB phase, the fast wind from the central star will interact with the material ejected during the AGB phase. The shock caused by this interaction will dissociate and heat the gas. This warm gas will cool through atomic fine structure lines of O I and the rotational lines of (newly formed) CO.
We present KAO observations of the CO J=9-8 and 12-11 lines in the C-star IRC+10216; these are the highest rotational transitions so far observed in any AGB star. A model is developed for the CO emission by IRC+10216 in all the observed CO rotational transitions.
We have developed a radiative transfer model of the dust and gas envelopes around late-type stars. The gas kinetic temperature for each star is calculated by solving equations of motion and the energy balance simultaneously. The main processes include viscous heating and adiabatic and radiative cooling. Heating is dominated by viscosity as the grains stream outward through the gas, with some contribution in oxygen-rich stars by near-IR pumping of H2O followed by collisional de-excitation in the inner envelope. For O-rich stars, rotational H2O cooling is a dominant mechanism in the middle part of the envelope, with CO cooling being less significant. We have applied our model to three well-studied oxygen-rich red giant stars. The three stars cover a wide range of mass-loss rates, and hence they have different temperature structures. The derived temperature structures are used in calculating CO line profiles for these objects. Comparison of the dust and gas mass-loss rates suggests that mass-loss rates are not constant during the asymptotic giant branch phase. In particular, the results show that the low CO 1-0 antenna temperatures of OH/IR stars reflect an earlier phase of much lower mass-loss rate.
We present 10 micrometer images of the multiple T Tauri star system UZ Tau taken with the Berkeley mid-infrared array camera at the United Kingdom Infrared Telescope (UKIRT) and evidence that UZ Tau E and W are a common proper pair. The mid-infrared emission is resolved for the first time into the two components UZ Tau E and UZ Tau W. The mid-infrared excess deduced for UZ Tau W appears to be much lower than that observed for UZ Tau E. This excess emission is consistent with an optically thin circumstellar disk in the case of UZ Tau W, whereas UZ Tau E's excess is consistent with an optically thick disk. We suggest that the close binary star pair in UZ Tau W is responsible for the observed difference between UZ Tau E and W's mid-infrared excess. In the proposed model, the binary star interacts with the local circumstellar disk environment and clears out much of the material inside its orbital radius (approximately 50 AU). As a result, the hot dust, observed at mid-infrared wavelengths, in UZ Tau W is suppressed compared to its 'wider' companion UZ Tau E. This scenario can also plausibly account for differences observed in UZ Tau E's and UZ Tau W's optical line strengths and profiles.
We present ground based observations of AGB stars for CO J=6-5; 4-3; 3-2; 2-1 and 1-0 lines. Radiative transfer modelling of these lines suggests that the stars do not lose mass continuously, but undergo discrete bursts of significant mass loss. Ratio of brightness of different fines show a wide range of behaviours in different stars, with higher transition lines probing deeper into the circumstellar envelope. We also obtained higher CO rotational transitions for post-AGB stars at J=22-21; 17-16 and 13-12. These lines can be used as probes for the warm medium at the inner part of the envelope. From these results, we can trace the history of mass loss in these stars.
We present ground-based spectra in the 10 and 20 micron atmospheric windows of the deeply embedded protostar GL 2136. These reveal narrow absorption features at 9.7 and 8.9 microns, which we ascribe to the CO-stretch and CH3 rock (respectively) of solid methanol in grain mantles. The peak position of the 9.7 micron band implies that methanol is an important ice mantle component. However, the CH3OH/H2O abundance ratio derived from the observed column densities is only 0.1. This discrepancy suggests that the solid methanol and water ice are located in independent grain components. These independent components may reflect chemical differentiation during grain mantle formation and/or partial outgassing close to the protostar.
Ground-based 7.4-24-micron spectra of two post-AGB objects, HD 161796 and HD 179821, are reported, and they are compared to those of other preplanetary nebulae. HD 161796 and HD 17982 show emission features at 10-12 microns and at 10 microns, and they exhibit a very rapid increase in flux between 13 and 15.5 microns. In view of the O-rich photosphere of HD 161796 and the presence of OH maser emission around all three objects, these features are ascribed to various oxides. The observed spectral features are quite different from the canonical silicate features observed in most O-rich giants. It is argued that HD 161796 and the bipolar nebulae Roberts 22 and NGC 6302 have all undergone the third dredge-up, with most of the dredged-up carbon having been converted to nitrogen by envelope-burning. It is concluded that carbon-rich grain material, produced during the interval between the end of the third dredge-up and the moment when envelope burning finally reduced the C/O ratio below unity again, could be responsible for the UIR bands now being excited in Roberts 22 and NGC 6302.
The 10-micron spectrum of the K5V star SAO 179815 (= HD 98800) is presented, and conclusively demonstrates the presence of small silicate dust grains around this star. The 9.7-micron silicate dust feature is unusually broad and shallow in this system. This, together with the slow fall-off of flux at longer wavelengths, constrains the size and density distributions of dust grains in models of the disk. It is found that there must be a significant population of small grains, as well as a population of large grains in order to explain all the observed properties of the disk.