Dissertation
X-ray lines: the high-energy signatures of massive-star wind dynamics
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Autumn 2023
DOI: 10.25820/etd.006920
Abstract
The extreme luminosities of massive stars drive hypersonic winds far denser than the solar wind, but they do so in a violent and nonsteady way that produces shocks and X-ray-hot gas via mechanisms that are not well understood. The goal of this thesis is to test long held assumptions in order to better understand the dynamics of the winds. I do this by using high resolution X-ray observations of the O supergiant $\zeta$ Puppis, the canonical massive star, to probe the shocked plasma's radial distribution and equilibrium state and developing new theoretical line profile models, one to test the underlying paradigm of the shock generation and another to capture the dynamics more directly.
The choppy nature of massive star winds have inherent instabilities that cause differential acceleration, leading to strong plasma shocks that heat small portions of the wind to X-ray emitting temperatures. When the gas is heated, it is assumed to reach an equilibrium plasma state known as collisional ionization equilibrium. This plasma state simplifies the problem of characterizing the X-ray emission by removing the requirement for an electron density in the calculation and requiring only a single temperature. For a plasma to reach this state, an optimal gas density is needed such that collisions are neither too frequent nor too infrequent. The shocks are thought to occur deep within the wind where the densities are near optimal, which also places them in a region where the hot gas cools more efficiently through radiative cooling by metal ions instead of adiabatic expansion. This deep-seated emission means that the observed lines contain information on both the wind dynamics leading up to the shock and the amount of material within the wind due reabsorption as it traverses the wind column.
Two tests were applied that focused on different aspects of these winds involving the use of observed X-ray emission from the winds. The first was whether that the X-ray emitting gas reaches collisional ionization equilibrium. If the plasma is not in this state, then assumed simplifications are not possible and the information in the X-rays require more detailed analysis to extract. The second focused on the boundary where the wind is initiated to see whether the wind shocks could be precisely modelled from the inhomogeneities being seeded by variable conditions at the base instead of generated from inherent instabilities. At the same time, an additional test was conducted on the efficacy of X-rays to be able to constrain the mass-loss rate of the star. The over-arching question being addressed is whether choices in the heating model can affect the diagnostic potential of the X-ray emission.
Recent theoretical considerations have suggested that the length scales for collisions is too great for the shocked gas to equilibrate before it mixes with the surrounding cold gas, putting impetus on the need to test the plasma state. A plasma in collisional ionization equilibrium has a predictable amount of flux within each emission line, so by comparing these values to measured satellite line and their associated resonance line flux, the equilibrium condition can be tested. Satellite lines are adept for this because the ratio of their flux to the resonance line flux is only temperature dependent, keeping the number assumptions that can compromise the results to a minimum. Using data collected by the \textit{Chandra} X-ray Observatory on $\zeta$ Puppis, I found that the measured and predicted flux ratios are statistically consistent, affirming that the plasma does reach the state of collisional ionization equilibrium.
For the second test, a line profile model is developed from assumed variable surface boundary conditions and fit to the same $\zeta$ Puppis data. I found to that this new model fit the data equally well as models built from the inherent instabilities, showing that different dynamical language can describe the same winds. An important parameter in this model is a fiducial optical depth that parameterizes the wind column depth and thereby the mass-loss rate, first developed by \citet{Cohen10}. My results confirm that their method is independent of the heating model used in defining the line profile model used for constraining of the mass-loss rate. However, while my results also confirm that the emissions lines of $\zeta$ Puppis, as measured by the \textit{Chandra}, have changed between subsequent observations, the source of the change is not in the mass-loss rate as \citet{Cohen20} concluded. They can alternatively be explained by changes in the distribution of the shocks within the wind, a less radical conclusion yet still raises further questions as to why.
The final portion of this thesis attempts to derive a model of the dynamics in a consistent method while also using simple language to connect model parameters to observables. The basis of the model is the analytic solution to the problem of Herbig-Haro Jets derived by \citet{Canto2000}. In deriving this new model, I generalize the jet solution to include flows with a prescribed non-zero acceleration. An example of line profiles generated from our model for the case of sinusoidal velocity and mass ejection variations are given as well. The example profiles show the expected shape of massive star X-ray emission lines and interesting but complicated trends with the model parameters. This establishes the possibility that observed X-rays could be a result of temporal variations seeded at the wind base, rather than purely generated intrinsically within the wind volume, and can be described via a quantitative language that connects with the physical attributes of those variations, consistently with the downstream momentum-conserving nature of shocked radial flows.
Details
- Title: Subtitle
- X-ray lines: the high-energy signatures of massive-star wind dynamics
- Creators
- Sean Jared Gunderson
- Contributors
- Kenneth Gayley (Advisor)Casey DeRoo (Committee Member)Keri Hoadley (Committee Member)David Huenemoerder (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Physics
- Date degree season
- Autumn 2023
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.006920
- Number of pages
- xvi, 149 pages
- Copyright
- Copyright 2023 Sean Jared Gunderson
- Language
- English
- Date submitted
- 11/20/2023
- Description illustrations
- illustrations, tables, graphs
- Description bibliographic
- Includes bibliographical references (pages 122-132).
- Public Abstract (ETD)
- The flow of material coming off a massive star is a chaotic stream of plasma clumps that accelerate outward at different rates due to their unstable nature. These winds differ from the Sun’s wind by the fact that massive stars are incredibly luminous, indeed so luminous that their radiation actually drives the wind in the same way as a “light sail” in science fiction. This light sailing is neither leisurely nor smooth for the gas because the way light couples to matter tends to magnify small variations because the coupling is via very Doppler shift sensitive line opacity. The winds reach hypersonic speeds that lead to clumps colliding and heating up to temperatures hot enough to emit X-rays, like what a doctor uses to image bones, but the how and why this happens is still unclear. My thesis tests theories previously developed to explain the violent dynamics using newly identified features in the X-ray observations of the well studied massive star ζ Puppis and creating new models that explain the unstable nature as being encoded from the conditions where they are launched rather than appearing later. I found that our theories of the heating are accurate, but the dynamical motion can be attributed to boundary variations influencing the clumps launching while describing the same wind. Finally, I used these boundary variation principles to develop a new model that more directly captures the wind’s dynamics with a useful language to connect with observationally measurable quantities.
- Academic Unit
- Physics and Astronomy
- Record Identifier
- 9984546750302771
Metrics
2 File views/ downloads
25 Record Views