Thesis introduction and summary
Where do we come from? One of the fundamental goals in all science is to understand the origins of life on Earth. The basic particles that make up life on earth, as well as the earth itself, the other planets and the sun, were created at the beginning of the universe, but how did they become what they are today? With the exception of hydrogen, some helium and some lithium, all the atoms in the universe were created by stars, through nucleosynthesis, where hydrogen atoms are processed to make other elements. These elements are then ejected from the stars in various ways, either through relatively gentle mass-loss in the case of low mass stars or explosively in the case of high mass stars. These new atoms form dust and molecules in space and are then incorporated into new stellar systems, planets and life. Hence, we are all made of stardust.
One of the sites of dust formation is in the atmospheres of asymptotic giant branch (AGB) stars (other sources include novae and supernovae). The aim of this work is to investigate the nature of the dust forming in these circumstellar regions, and in particular the mineralogical composition of the dust. This involves an understanding of the nucleosynthetic processes that create the elements from which the dust forms, and of the chemistry that determines the specific type of dust that forms. The method of observing these dust grains uses spectroscopy1, which enables their identification through the recognition of the spectroscopic signature of a specific mineral. We can use spectroscopy of terrestrial samples in the laboratory to identify these signatures for individual minerals, but we must also understand the way in which electromagnetic radiation (light) from the star interacts with surrounding dust to produce the observed stellar spectra.
The study of meteorites has provided useful information on the nature of the dust and molecules forming in the circumstellar regions around AGB stars. Isotopic studies have shown that meteorites contain presolar grains that have survived intact since their formation around such stars. The results of the spectroscopic studies of stardust should be be related to, or even constrained by, presolar grains in meteorites
Most stars whose main-sequence mass is less than about 8 solar masses evolve into a phase known as the Asymptotic Giant Branch or AGB. These stars have low effective temperatures (< 3000K) and high luminosity (> 3000 solar luminosities). These ``low-to-intermediate mass'' stars build a core of degenerate electron gas, with a helium-burning shell around it and surrounding that, a mantle of hydrogen2. It is the hydrogen-burning shell that is responsible for their high luminosity. During the AGB phase stars begin to expel matter. This matter ejection process leads to the formation of a circumstellar envelope around the star. It is this circumstellar envelope that is of interest in a discussion about stardust.
The basic scenario is this: an AGB star has a spherically symmetric gas outflow. Dust particles form in the circumstellar envelope at some distance from the star. The precise nature of the circumstellar dust grains varies from star to star. The dust's evolution and physical properties are dominated by the star. For example, the temperature of the dust grains is dependent on the nature of the grains and the central star. The chemical composition of the grains depends on the nature of the gas being ejected from the star. In turn the chemical composition of the ejected gas is controlled by the nucleosynthesis within the star. The chemistry around a star is usually either dominated by carbon or by oxygen. The reason for this is the remarkable stability of the carbon monoxide (CO) molecule. The molecule forms very easily and is very stable. Therefore, depending on the ratio of carbon-to-oxygen atoms around a star, one of the species will be completely locked into CO molecules, leaving the other to form dust and molecules. If the carbon-to-oxygen ratio (C/O) is less than unity, all the carbon atoms will be locked into CO, leaving the excess oxygen atoms to dominate the chemistry. Similarly, if C/O > 1, there will be an excess of carbon dominating the chemistry. The rare exception to this is when C/O=1 and the carbon and oxygen atoms are equally depleted, leaving other atoms (e.g. nitrogen, sulphur) to form the majority of the dust and molecules.
The dust grains around a central star absorb light from the star and re-emit this light in the infrared region. The exact nature of the re-emitted light depends on the nature of the incident light as well as the chemical and physical composition of the dust and molecules. Therefore, we can use this infrared emission to detect and investigate circumstellar dust shells. The infrared emission from the dust is detected in the form of both continuum emission (infrared excess) and discrete spectral features due to individual dust species. These spectral features have been investigated by many people (e.g. see reviews by Woolf 1973, Woolf & Ney 1969, Bode 1988 and also sections 6.1, 6.2, 7.1, 8.1). The quest to more accurately identify these features incorporates the use of many laboratory mineral spectra, an understanding of how these minerals interact with the incident light, and models of the condensation sequences of dust expected in these regions.
The widely observed 9.7 µm feature is generally accepted to be due to Si-O bond stretching in silicate dust (assumed to be mostly iron and magnesium silicates). The profile of the feature varies from star to star and is thought to be due to variations in the exact nature of the silicates (e.g. whether crystalline or amorphous, which cations are present, what impurities are present, etc.). Silicates also have a feature at about 18 µm due to the O-Si-O bond bending.
There is another relatively common spectral feature at about 11.3 µm seen in carbon-rich AGB star spectra, which has been attributed to silicon carbide (SiC). The silicon carbide feature has been the subject of some debate (Friedemann et al. 1981, Borghesi et al. 1985, Pegourie 1988, Groenewegen 1995, Speck et al. 1997a,b, Papoular et al. 1998), not least because presolar SiC grains have been found in meteorites with isotopic compositions suggestive of AGB star origins. The data from these meteoritic grains has disagreed with the astronomical observations of the 11.3 µm feature, and this problem is further addressed in this thesis.
Another set of features seen in the spectra of many astronomical objects are the so-called unidentified infrared (UIR) bands. These are a set of narrow features found in the infrared region of the spectrum. Their relative strengths vary from source to source. These bands have been attributed to some form of hydrogenated carbon, although whether it is in the form of molecules or dust grains is still under debate. Given that they would need a carbon-rich environment in which to form, carbon stars have been suggested as a possible source of such species. Unfortunately the UIR bands need a source of ultraviolet (UV) radiation to be produced, which is something carbon-rich AGB stars have a distinct lack of, and therefore these dust grains/molecules would not be seen even if they were present around such stars. However, some carbons stars can be found in binary systems with hot companion stars which can provide the UV radiation needed to produce the UIR bands. Therefore, these bands may be observable in the spectra of these carbon stars.
The work presented in this thesis is broadly split into two sections based on the chemistry: carbon-rich and oxygen-rich. In chapter 2, the basic evolution and nucleosynthesis of AGB stars is discussed. How these objects evolve from stars with oxygen-rich atmospheres to stars with carbon-rich atmospheres is outlined. Chapter 3 gives an introduction to meteorites and in particular presolar grains. The aim of the chapter is to help the reader understand how the isotopic analysis of presolar grains has lead to a better knowledge of their supposed origins, and how this relates to observational spectroscopy3.
As discussed above, the use of spectroscopy requires an intimate knowledge of how matter interacts with light. Chapter 4 contains a basic discussion of the principles of the optical properties of matter. During this work, an error in previous laboratory spectra was uncovered. This is discussed and has implications for chapter 6. Following on from the discussion of optical properties of matter, essential to the astronomical spectroscopist is a good selection of laboratory spectra with which to compare observations. Therefore, chapter 5 contains a catalogue of mineral spectra of use to astronomers, with a discussion of the similarities and differences between the spectra and how they relate to the theoretical dust condensation sequences around AGB stars.
Chapter 6 covers the interpretation of the 11. µm feature in the spectra of carbon-rich AGB stars. The implications of the findings in chapter 4 regarding errors in laboratory spectra are also discussed and how they lead to the reconciliation of astronomical and meteoritic data on silicon carbide. This is followed in chapter 7 by further studies of carbon-rich AGB star spectra, with respect to UIR bands. The observation of a UIR band in the spectra of two carbon-rich stars with hot star companions is discussed. There is also a brief discussion of the possibility of diamond dust forming around these stars, prompted by the presolar diamonds found in meteorites.
Chapter 8 investigates the spectra of oxygen-rich AGB stars. These stars present a much wider range of spectral features and theoretical condensation models suggest there are many possible minerals forming in the circumstellar regions around these stars. Oxygen-rich AGB stars present a much more complicated situation than the carbon-rich stars. The basic interpretation of the 9.7 µm feature as a silicate is investigated and an attempt is made to constrain the identification further. The discovery of a new feature at 9.2 µm, attributed to silicon dioxide dust is discussed. Some forms of silicon dioxide also have a spectral feature at 12.5-13. µm. A similar featyre in the O-rich AGB star spectra, previously identified as an corundum (Al2O3) feature, may also be attributable to silicon dioxide.
Finally, chapter 9 summarises the results from the previous chapters and discusses future research that should be undertaken to benefit from the advances made in this work.
1The study of objects by analysing the way they reflect, transmit and absorb electromagnetic radiation.
2The exact structure of the interior of the star depends on the initial mass and the metallicity. This is discussed in more detail chapter 2.
3Whilst trying to explain that I study dust around stars, an archaeologist friend asked ``But how do you get the dust down?''. The answer is in meteorites
 Angela's Home Page |
 speckan@missouri.edu |
 Angela's Research Page |