research | teaching | links
Research

    Classical Novae
    A classical nova is thought to occur in a binary system consisting of a white dwarf accreting from a main sequence (or late type giant) via Roche Lobe overflow. The accreted material builds up as a layer on the surface of the white dwarf until the base temperature increases to the point when a thermonuclear runaway occurs, dominated by CNO reactions. This shell burning continues at Edddington luminosity for the white dwarf. Material is ejected at speeds from a few hundred to a thousand km/s. The shell starts as an optically thick fireball and expands and cools to be optically thin. Novae are classified by the speed at which they decrease in brightness. Classical novae are thought to recur every 10,000 years or so.

    Some novae evidence a drop in optical/UV light with a corresponding increase in IR emission. This change represents the production of dust, which reprocesses the optical/UV light into infrared. It seems that fast novae do not, on average, produce dust, while slow novae can evidence optically thick or thin dust emission or none at all. However, a unified model of dust production does not exist to explain these difference scenarios.

    Images of classicl novae. Top row, left to right: DQ Her (Slavin 1995), HR Del (Harman & O'Brien 2003), T Aur (Slavin 1995). Bottom row, left to right: GK Per (NOAO), FH Ser (Gill & O'Brien 2000), T Pyx (Shara et al. 1997).
    My thesis is based on OIR photometry and optical spectroscopy of the slow Nova Cen 1991. Initial results indicate Nova Cen produced dust but also has simultaneous contamination by optical light, indicating less than uniform ejecta. Evidence exists for clumpy ejecta in images of the GK Per and the light echo from V838 Mon, and explains how dust can form in an environment which would otherwise destroy the grains. Inside a dense clump ejecta can cool to the estimated <2000 K required to condense dust grains, as well as protecting the growing grains from radiation. I intend to model Nova Cen with clumpy ejecta using DIRTY. This presents a solution to the variable dust formation observed, with detection of dust depends on a line-of-sight through a clump.

    X-ray Binaries
    An X-ray binary is a system of two stars where one is a compact object accreting material from the other, and they are named for their large X-ray luminosities. The accretor is more evolved and has exhausted its nuclear fuel, existing as a stellar cinder like a neutron star or a black hole. The donor is usually a main sequence, giant, or supergiant star but can also be a white dwarf. In a High Mass X-ray Binary (HMXB) the primary accretes matter from the secondary's stellar wind. The secondary is usually an early type (young and massive) star with a strong stellar wind. In a Low Mass X-ray Binary (LMXB) the secondary is larger than its Roche Lobe, the surface of gravitational equipotential inside of which material is bound to the star. Material overflowing the Roche Lobe is attracted by the primary, again a compact object.


    Artist rendition of an X-ray binary. (Image credit: Astronomical Illustrations and Space Art by Fahad Sulehria)
    In both types of X-ray binaries material attracted to the accretor forms (by conservation of angular momentum) an accretion disk around it. Viscous processes in the disk serves to heat the material in the disk to millions of degrees (emitting X-rays as a black body) and redistribute angular momentum such that material migrates to the inner part of the disk. Material loses gravitational potential energy as it falls toward the accretor and emits X-rays.

    X-ray binaries have multiple sources of light at various wavelengths. The accretion disk itself usually overwhelms all other sources from visible to X-rays. The disk is thought to have a corona of tenuous hot gas which can reprocess (absorb and reemit) the accretion disk's X-rays as IR radiation. In LMXBs the donor would only contribute significantly to the IR, but in HMXBs it would be much more hot and luminous and contribute to visible and UV light. X-ray binaries are thought to have synchrotron jets (usually) perpendicular to the accretion disk, which are streams of relativistic plasma radiating across the spectrum from X-rays to radio. Due to the other competing bright sources of light, these jets are usually only observed in radio and hard X-rays and only in black hole systems. Synchrotron jets have been observed in X-ray binaries such as GRS 1915+105 (Fender et al. 1997), GX 339-4 (Corbel & Fender 2002), and XRB 4U 0614+091 (Migliari et al. 2006).

    GX 17+2 is a LMXB in the constellation Serpens (near the galactic center). This system is also a Z source, meaning that the neutron star is accreting at or near the Eddington rate (accreting almost to the point where the energy released blows away the accreting material). Z sources are so named for the pattern they trace out on a X-ray color-color diagram, thought to relate to the ratio of the mass accretion rate and the average mass accretion rate (Hasinger & van der Klis 1989). GX 17+2 is one of the most active of the eight known Z sources. Its X-ray and radio activity seems correlated with radio brightness increasing from the NB to the HB (Penninx 1988, Migliari & Fender 2006), and it evidences the largest amplitude of rapid radio variations (Penninx 1989). Migliari & Fender (2006) found a correlation between radio and X-ray activity in Z sources in the hard X-ray state, where rapid state changes are associated with bright, optically thick, transient radio events (flares).

    K band light curves for four IR brightening events, defined as when the blend of GX 17+2 and the field star NP Ser is brighter than the resolved measurement of NP Ser itself, indicated by the line at 14.5.

    GX 17+2 demonstrates IR variations of an amazing 3 magnitudes in the K band (2.2 microns), which activity had never been seen in this source nor in any of the other Z sources. Intrigued by this unusual behavior, we studied this source for three years and accumulated the largest dataset of time-resolved IR light curves. We have detected 6 IR brightening events. Based on our data and that of Callanan et al. (2002), we have determined the IR brightening events to recur on a period of about 3 days. Futher, the light curves are not uniform in character, some having a sharp peak and others displaying a long plateau, or in peak magnitude.

    These IR events are not associated with X-ray activity, implying that they are not caused by accretion disk events or coronal reprocessing. GX 17+2's IR variations are not due to eclipses, for the system brightens during the course of the event, which is fairly short in relation to the period. They cannot be due to mass ejection events (flares), for flares are stochastic in nature and our observed behavior is periodic.

    We suspect GX 17+2 posses a precessing IR synchrotron jet. The 3 day period is a curiosity, for the standard model implies a stochastic jet. It is possible to explain these observations as a precessing synchrotron jet, as seen in BL Lacs (supermassive black holes at the centers of some galaxies), with only the inner disk precessing to allow it to have such a short period. We presume that we observe a chord of the conical jet as it precesses across our line of sight. This explains both the periodic and variable aspect of the observed IR brightening events. Current theoretical models predict such IR synchrotron radiation but also predict it will only be brighter than the reprocessed accretion disk corona emission for short-period systems (Migilari, et al. 2006). GX 17+2's orbital parameters are unknown, however.

    This work has been published in ApJ Letters, "A Possible Period for the K-band Brightening Episodes of GX 17+2".

Teaching

    I am not teaching any lab sections at this time. I have been a TA for an undergraduate introduction to astronomy class (which amounts to running the lab portion of the course), a history of spaceflight course, and a course and life in the universe.

Link Hoard

research | teaching | links