Digging for Substellar Objects in the Stellar Graveyard

Open Access
Debes, IV, John Henry
Graduate Program:
Astronomy and Astrophysics
Doctor of Philosophy
Document Type:
Date of Defense:
March 29, 2005
Committee Members:
  • Steinn Sigurdsson, Committee Chair
  • Michael Eracleous, Committee Chair
  • James Kasting, Committee Member
  • Alexander Wolszczan, Committee Member
  • Lawrence William Ramsey, Committee Member
  • white dwarfs
  • extrasolar planets
  • brown dwarfs
  • high contrast imaging
White dwarfs, the endpoint of stellar evolution for stars with with mass $<$ 8$Msun$, possess several attributes favorable for studying planet and brown dwarf formation around stars with primordial masses $>$ 1 $Msun$. This thesis explores the consequences of post-main-sequence evolution on the dynamics of a planetary system and the observational signatures that arise from such evolution. These signatures are then specifically tested with a direct imaging survey of nearby white dwarfs. Finally, new techniques for high contrast imaging are discussed and placed in the context of further searches for planets and brown dwarfs in the stellar graveyard. While planets closer than $sim$ 5~AU will most likely not survive the post-main sequence evolution of its parent star, any planet with semimajor axis $>$ 5~AU will survive, and its semimajor axis will increase as the central star loses mass. The stability of adjacent orbits to mutual planet-planet perturbations depends on the ratio of the planet mass to the central star's mass, and I demonstrate that some planets in previously stable orbits around a star undergoing mass loss will become unstable. I show that when mass loss is slow, systems of two planets that are marginally stable can become unstable to close encounters with each other, while for three planets the timescale for close encounters decreases significantly with increasing mass ratio. Close encounters can lead to collisions of planets and new orbits that perturb surviving planetesimals into white dwarf grazing orbits. Perturbed planetesimals can create pollution of the white dwarf's surface and circumstellar dust disks which can be observational signatures of planetary systems If pollution of a white dwarf's atmosphere is caused by relic planetary systems, any white dwarf with photospheric absorption due to metals can be searched for substellar companions. Hydrogen white dwarfs with metal absorption, so called DAZ white dwarfs, are hard to explain by simple ISM accretion, and present an opportunity to test the observational signatures of unstable planetary systems. Additionally, field white dwarfs can be searched for substellar companions as well. I conducted a Hubble Space Telescope (HST) and Canada France Hawaii Telescope survey of nearby white dwarfs for substellar objects. A total of 8 DAZ white dwarfs and a total of 12 field dwarfs with distances of $<$ 50 pc had high contrast and high spatial resolution images taken with a combination of ground based and space based observations to search for candidate substellar objects at separations $ltorder$10arcsec away. Limits to unresolved companions are derived through analysis of 2MASS photometry of the white dwarfs compared to expected fluxes based on the WDs effective temperature, distance, and gravity. From my survey I find several candidate companions, which have been or will be followed up with second epoch observations. I find that my observations are sensitive to planetary companions around some targets with M $gtorder$6 M$_{Jup}$, and sensitive to companions with M $>$ 18 M$_{Jup}$ for all but two of my targets. The lack of significant near infrared excesses for my targets limits any kind of unresolved companions present to be substellar. In light of these results I make several comments on the possibility of determining the origin of metals in the atmospheres of white dwarfs and the frequency of substellar objects in orbit around intermediate mass stars. The search for planetary companions to stars requires further development of high contrast imaging techniques. This thesis studies Gaussian aperture pupil masks (GAPMs) which in theory can achieve the contrast requisite for directly imaging an extrasolar planet around a nearby solar type star. I outline the process of designing, fabricating, and testing a GAPM for use on current telescopes and specifically the Penn State near-IR Imager and Spectrograph (PIRIS) at the Mt. Wilson 100$^{primeprime}$ telescope. I find that observations with a prototype are quite successful, achieving a contrast similar to a traditional Lyot coronagraph without blocking any light from a central object and useful for finding faint companions to nearby young solar analogues. In the lab I can reproduce the expected PSF reasonably well and with a single aperture design which achieves $sim4 imes 10^{-5}$ contrast at 10$lambda/D$. I find that small inaccuracies in the mask fabrication process and insufficient correction of the atmosphere contribute the most degradation to contrast at these levels. Additionally I present the first laboratory experiments using a notch-filter mask, a coronagraphic image mask that can produce infinite dynamic range in an ideal Lyot coronagraph according to scalar diffraction theory. I fabricated the first notch-filter mask prototype with 0.25 micron precision using an e-beam lithography machine. My initial optical tests show that the prototype masks generate contrast levels of 10$^{-5}$ at 3$lambda/D$ and 10$^{-6}$ at $sim 8 lambda/D$, with a throughput of 27\%. I speculate on the ``as-is' performance of such a mask in the Hubble Space Telescope and for white dwarf targets.