Workshop on
Planetary Formation in Binary Environment

Stony Brook, USA
16 - 18 June 1996

Organised by Andrea Ghez and Michael Simon

ABSTRACTS FOR THE INVITED TALKS

HTML version compiled for the Extrasolar Planets Encyclopedia by J. Schneider

Interaction of planets, superplanets, and binary stars with disks

Binaries and planets are born in, and have ample opportunity to interact with, protostellar and protoplanetary disks in YSOs, PMS stars, and solar nebulae. Since most even if not all of that interaction occurs via gravity, theorists learned to deal with various aspects of gravitational coupling, and concluded that resonant excitation of density waves plays a major role in shaping the disk's response to a planet/star, and the feedback of the disturbed disk on the perturber. I review the results of applying theoretical concepts to problems involving: (i) small planets and planetesimals during planet accumulation, (ii) Jovian planets, which are able to open disk gaps, as well as (iii) super-Jovian planets (superplanets) of both the circular and eccentric varieties. Finally, I review 1-dimensional and multi-dimensional approaches to binary-disk interaction, and discuss mass flows through circumbinary gaps. If that last phenomenon also occurs in planet-disk systems in nature, it may have new important consequences for the termination of gas accretion onto giant planets (i.e., their final masses), as well as the final shapes and sizes of their orbits. A larger variety of planets including superplanets may form in disks than realized heretofore.

Formation of Brown Dwarfs and Giant Planets

Brown dwarfs and giant planets are commonly distinguished by the formation process. The relatively high mass and high orbital eccentricity of the companion to 70 Vir has led to speculations that it is a brown dwarf rather than a giant planet. Another recent discovery is the brown-dwarf companion to GL229. Such brown-dwarf companions to main-sequence stars can be formed in the same way as normal stars, by collapse and fragmentation of the rotating cores of molecular clouds. The main limitation is that the mass must be greater than about 0.007 M_sol, which is the minimum Jeans mass along the temperature-density path of the collapsing fragment. The alternatives for the formation of GL229B, and possibly for the companion to 70 Vir, are 1) gravitational instability in a disk, leading to a subcondensation; 2) direct fragmentation of a collapsing cloud, leading to a system of small mass ratio; or 3) induced fragmentation of low-mass objects in a multiple system, followed by capture of a low-mass fragment by one of the major fragments. The merits of these various scenarios will be discussed.
The observed properties of giant planets constrain theories on their origin. In the case of the solar system, the giant planets were formed from both a gaseous and a solid component, with the heavy elements enriched relative to Solar composition. Some of the heavy elements may form a solid core, and the rest is dissolved in the gaseous envelope. Other crucial observations are the measured internal heat fluxes and the presence of satellite systems. In the case of extrasolar giant planets, little is known except their masses (M sin i) and orbital properties.
The standard picture of the formation of a giant planet involves several stages: 1) accretion of solid particles forms a solid core of a few earth masses, accompanied by a very low mass gaseous envelope; 2) further accretion of both gas and solids results in the mass of the envelope increasing faster than that of the core until a crossover mass is reached; 3) runaway accretion of gas occurs until 4) accretion is terminated by tidal truncation or the dissipation of the nebula; 5) the planet contracts and cools at constant mass to its present state, with a relatively high and possibly observable luminosity at the earlier stages. The main problems with this model are 1) the formation times for the solid cores may be too long compared with estimated disk lifetimes; 2) the model does not explain Uranus and Neptune, which evidently have not evolved to the crossover point; and 3) the model does not explain the close orbital separations of two of the extrasolar giant planets or the high mass of the companion to 70 Vir. Recent calculations will be presented which address some of these problems.

A survey of binary systems among intermediate-mass pre-main sequence stars

We have observed 44 objects randomly selected from The et al.'s (1995) list of Herbig Ae-Be candidates by means of diffraction-limited imaging in the visible and near-infrared ranges (adaptive optics, speckle interferometry). We detected 13 binaries with separations between 0.13 and 3.1 arcsec, which yields a binary frequency of 30% in this sample. I shall present the spectral energy distributions of the primaries and secondaries, covering the visible and near-IR domains, and discuss the nature and properties of the companions. For a few systems, we additionally obtained low-resolution spectra of the companion which will be presented.

Binary Surveys: Constraining the Number of Planetary Systems

We have surveyed 195 X-ray selected pre-main sequence stars, most of which are weak-line T Tauri stars, for visual binaries with separations between 0.8 and 12'' (i.e. 120 AU to 1800 AU at a distance of 150 pc). The T Tauri stars had been identified in the course of optical follow-up observations of sources from the ROSAT All Sky Survey associated with star forming regions. The areas surveyed include the T associations of Chamaeleon and Lupus as well as Upper Scorpius, the latter being part of the Scorpius Centaurus OB association (Sco OB 2). The binary frequency in Scorpius Centaurus is of special interest as up to 80% of all field stars might originate in OB associations.
A comparison of the X-ray luminosity of the stars with the observed binary frequency indicates that the ROSAT All Sky Survey is biased towards (with ROSAT unresolved) binaries. If we correct for this bias, we find that the overall binary frequency among T Tauri stars in a range of separations between 120 and 1800 AU is in agreement with the binary frequency observed among main sequence stars in the solar neighborhood.
We conclude that binarity is established very early in stellar evolution and that the orbital parameters of wide binaries (sep. > 120 AU) remain virtually unchanged during their pre-main-sequence evolution.
If one assumes that the distribution of orbital parameters among pre-main-sequence stars is the same as among main-sequence stars (cf. Duquennoy & Mayor 1991) over the whole range of separations, ~ 70% of all pre-main-sequence binaries should have separations between 1 AU and 400 AU. In this range of separations, circumstellar disks in binaries will be strongly disturbed by the presence of the companion, and hence might not have the potential of forming planets. On the other hand, 30% of all binaries have separations larger than 400 AU or smaller than 1 AU and thus might eventually form planets either out of their individual circumstellar disks (like single stars) or out of circumbinary disks. For a binary frequency of 60% (100%), up to 58% (30%) of all stellar systems (single stars and binaries) could possess planets.

Orbital Migration of Giant Protoplanets

The recent discovery of an approximately Jovian mass planet orbiting in relatively close proximity to 51 Peg provides indirect evidence for the inward migration of giant protoplanets through planet-disk tidal effects. We produce hydrodynamic simulations of this tidal interaction, demonstrating protoplanets' ability to clear gaps in their disks and to spiral inward on an appropriate nebular timescale. As the protoplanet approaches the central star, the tidal force due to the star grows and eventually halts the inward migration. We produce models in which multiple giant planets evolve in such fashion within the same disk.

Multiple fragmentation in a centrally condensed protostar

Hydrodynamical calculations in three space dimensions of the collapse of an isothermal, centrally condensed, rotating 1 M sol protostellar cloud are presented. A numerical algorithm involving nested subgrids is used to resolve the region where fragmentation occurs in the central part of the protostar. A previous calculation by Boss, which produced a hierarchical multiple system, is evolved further, at comparable numerical resolution, and the end result is a binary, with more than half of the mass of the original cloud, whose orbital separation increases with time as a result of accretion of high-angular momentum material and as a result of merging with fragments that have formed farther out. Repeating the calculation with significantly higher resolution, we find that a sequence of binaries can be induced by fragmentation of circumbinary disks. The stability of the resulting multiple system is investigated using n-body calculations, which indicate that such a system would transform on a short time scale into a more stable hierarchical structure. The outermost and most massive binary which forms in the high-resolution run has properties similar to that of the binary found in the low-resolution calculation. Thus the basic outcome is shown to be independent of the numerical spatial resolution. The high-resolution run, in addition, leads to the formation of a system of smaller fragments, which might be important for the understanding of the origin of close binaries with low-mass components and of low mass single stars.
Additional calculations demonstrate that the final orbital parameters of the multiple systems are affected strongly by small initial perturbations, indicating that chaotic processes play an important role in determining the physical parameters of such systems.

Star-disc interactions and binary star formation

We discuss three projects that explore the connection between planet formation, circumstellar discs and binary star formation. Firstly we consider star formation in the context of small dense clusters, where interactions between circumstellar discs can both enhance the formation of binary stars (through star-disc capture) and possibly reduce the potential for planet formation as a result of disc destruction. It is shown, through detailed simulations, that although the destruction of the outer parts of discs during close encounters radically diminishes the potential for star-disc captures during subsequent encounters, the inner regions of the disc are relatively undisturbed, implying that planet formation is unlikely to be radically affected. A second project considers the formation of discs during the collapse and fragmentation of a molecular cloud core, exploring under what circumstances discs form around the primary and/or secondary components or else as a circumbinary disc. Again the implications for planet formation are discussed, as well as issues relating to the orbital evolution of the binary. Finally, we discuss work that considers the effect of a protoplanet (similar, perhaps, to the progenitor of 51 Peg) on the evolution of protostellar discs and the possible relationship between such objects and FU Orionis outbursts in young stars.

Disks Properties Derived from Millimeter Wave Interferometry

In the last few years, mm arrays have observed in continuum and in the first rotational lines of CO many disks around T Tauri stars belonging to the nearest star-forming regions, and the mm wave interferometry has become a powerful technique to retrieve the disk properties. Since mm arrays are now reaching spatial resolutions similar to optical or NIR telescopes, mm observations are really becoming complementary to optical and infrared data. Moreover mm interferometers still provide the unique solution to estimate the large scale kinematics of disks (down to a scale of ~ 100 AU) and they allow in some cases a detailed modeling which is a promising step toward the understanding of the formation and evolution of protosolar-like nebulae. After a brief introduction on the possibilities and limitations of mm techniques, I will mainly discuss through recent results the status of knowledge on protoplanetary disks coming from mm arrays.

Estimating Masses, Ages, and Accretion Rates among Pre-Main-Sequence Binaries

We must have reliable estimates for the ages and masses of the components of binaries in order to understand how these systems form and evolve. Unfortunately, the apparent luminosities of young stars with optically thick disks are difficult to measure directly because protoplanetary accretion disks emit strongly at both infrared and optical wavelengths. Uncertainties in the spectral types of the stars, their distances, and the reddening also complicate mass and age estimates. In this talk I will address these issues, consider whether or not components of young binaries appear to be coeval, and describe some of the techniques that can be used to determine whether or not a given binary component has an inner disk that accretes onto the star.

Long-Term Stability of Planets in Binary Systems

We investigate a simple question of celestial mechanics: in what regions of phase space near a binary system could planets persist for times on the order of age of the system? An extensive body of literature is dedicated to this question, but this literature has two major limitations. First, although binaries typically have eccentric orbits, most of the analytic results have been developed for the case of the binary in a circular orbit. Second, numerical investigations of stability have been limited to fairly short integrations. Although it is difficult to extend the available analytic results, faster computers and improved algorithms make it a straightforward task to improve the numerical results.
We employ the elliptic restricted three body problem to investigate orbital stability numerically. For the orbital integrations we used the symplectic mapping method of Wisdom and Holman (1991). For a range of values of the binary eccentricity and mass ratio, we investigate orbits either near one of the stars or well outside the pair. From the results, we develop empirical expressions for (1) the most distant orbit around one star or the other in which test particles survive the length of the integration, and (2) the nearest orbit outside of the binary in which a test particle in a nearly circular orbit about the center of mass of the system would survive the full integration.
We also examine two specific systems in detail. The first is Alpha Centauri, a triple system composed of a close binary (23~AU, e=0.52) and a distant companion (~ 12,000 AU). Neglecting the companion and integrating test particles in the field of the binary, we find that initially circular prograde orbits in the plane of the binary within 3~AU of alpha Cen A survive the full length of the integration (currently ~ 10**5 orbits of the binary or ~ 10**7 years). Likewise, initially circular retrograde orbits in the plane of the binary within 4.4~AU of alpha Cen A survive for full integration. Orbits inclined 90 deg. to the binary must be within 0.18~AU to survive the full integration.
The second and final specific case we investigate is a system of interacting planets in the field of an eccentric binary pair. For initial conditions we simply take the sun and giant planets from our own solar system and add a solar mass companion. The companion has a prograde orbit in the invariable plane of the solar system with an eccentricity of 0.4. We check several initial semimajor axes for the companion. For initial semimajor axes of 150, 200, and 250~AU we find that Uranus and Neptune cross orbits within 30~million years. However, for an initial semimajor axis of 500~AU the system of planets survives for at least 1 billion years. This is roughly consistent with the results obtained from the elliptic restricted three body problem. Although our own system of planets might be different from the type of system that could form in the binary environment, this experiment is an important proof of concept.

Observations of Circumstellar and Circumbinary Disks: Where is the Mass?

Current observations cannot resolve disk material around nearby young binaries at scales smaller than a few tens of AU. Thus, in most binary systems we do not know where most of the disk mass lies, information which is crucial for determining what type of planets (if any) will form in these systems. I will discuss current observational constraints on the location of disk material in young binary systems, and in particular the division of material between circumstellar and circumbinary disks.
Young binaries with projected separations between 1 and 50-100 AU have lower millimeter fluxes than wider binaries or single stars, consistent with low disk masses due to close binaries clearing large gaps in their disks. However, two-thirds of the binaries in this separation range were detected by IRAS at 12, 25, and 60 microns. Thus, most binaries in this separation range clearly retain some circumstellar material. These circumstellar disks most likely have masses less than the minimum mass solar nebula, though they may have enough mass to form Earth-like planets.
Aperture-synthesis images of the young quadruple system UZ Tau at lambda = 1.3 mm with 1 '' resolution show that UZ Tau W, a 50 AU binary, has 1.3 mm emission which is a factor of four lower than that from UZ Tau E, a sub-AU spectroscopic binary 500 AU distant. UZ Tau E has a 150-AU radius circumbinary disk as traced by 12CO (2->1) emission. In contrast, UZ Tau W has unresolved continuum emission and no detectable CO emission, indicating that it has at least one circumstellar disk but no circumbinary disk. Since E and W have similar ages and stellar properties, differences in their disks must be attributed solely to the effects of the differing locations of their binary companions, and emphasizes the importance of not only multiplicity but specifically binary separation in determining disk properties.
While observations suggest that the closest binaries (those with separations less than 1 AU) may retain substantial circumbinary disks, these are also the binaries that may most strongly disrupt the terrestrial-planet-forming region of their disks. We have examined the spectral energy distributions (SEDs) of young spectroscopic binaries with infrared excesses to look for evidence of holes or gaps in their disks. Several show large dips in their SEDs which are consistent with completely cleared inner holes on the size scale of the binary separation. However, others show near-power-law infrared SEDs and strong accretion diagnostics. The SEDs of these binaries, which include strong 9.8 microns silicate emission features, are consistent with optically thin but not entirely evacuated regions near the binary orbits. This suggests that a close binary will not necessarily clear an evacuated central hole and that material may continue to flow from circumbinary to circumstellar disks.
This work done in collaboration with Robert D. Mathieu, David W. Koerner, and Gary A. Fuller.

Multiplicity of T Tauri Stars in Taurus after ROSAT

Recent studies of the nearby Taurus and Ophiuchus star forming regions have shown that the rate of binary and multiple systems among T~Tauri stars is significantly higher than among Main-Sequence stars. In fact, all or nearly all T~Tauri stars form in binary systems.
These studies concentrated on the population of ``classical'' T~Tauri stars (equivalent width of the H alpha emission line > 10 Angstr. However, follow-up observations of sources in the ROSAT All-sky Survey have shown that the ``weak-line'' T~Tauri stars outnumber the ``classical'' T~Tauri stars by at least a factor of eight.
In order to derive results representative for the whole population of young stars in Taurus, we carried out a survey of 76 weak-line T~Tauri stars in Taurus (which were discovered using ROSAT observations). We used NIR-speckle interferometry with the MAGIC camera in the K Band 2.2 microns at the 3.5 m telescope on Calar Alto, Spain.
Preliminary results are presented and discussed.

Discovery of a Cool, Brown Dwarf

For the past five years we have been conducting an optical coronagraphic search for faint companions to nearby stars. The Johns Hopkins University's Tip-Tilt Coronagraph has been used on the Palomar 60-inch. With this device, we have achieved a dynamic range of 13 mag. Each potential candidate is reobserved a year later. The criterion for companionship is that the candidate has the same proper motion as the parent star. Candidates thus identified are pursued with imaging and spectroscopy in the 1-2.5 micron, the near-IR band, on the Hale 200-inch. We report a common proper motion companion to Gliese 229 (=HD 42581, HDS 1827; spectral type M1V; M_V=9.3; 5.7 pc). The object, Gliese 229B, has a peculiar broad-band spectrum peaking at 1 micron. It is very red in the optical and blue in the near IR. We see strong first overtone absorption bands of methane, strikingly similar to that seen in the spectrum of sun light reflected by Jupiter. UKIRT spectroscopy has revealed bands that can be unambiguously attributed to water. Keck spectroscopy in the range 0.8-1.0 micron clearly show strong steam features. There are no strong bands from refractory elements (TiO, VO, FeH) that dominate the spectrum of the coolest stars. We argue that Gliese 229B is a brown dwarf. Tsuji's dust-free model of a 900 K brown dwarf, whose opacity is essentially due to water and methane, fits our broad-band data from 0.6 micron (Hubble Space Telescope) to 10 microns (Keck and Palomar). We conclude that Gliese 229B is the first unambiguous detection of a cool brown dwarf whose mass is well below the sub-stellar limit and most likely mass of ~ 30 M_Jup. The incidence of such companions is rare enough to make them uninteresting as candidates for dark matter.

Speculations Regarding Planetary Formation In Binary Star Systems

Modern theories of planet formation are based primarily on our own Solar System. These theories predict that most single stars should have planets in orbit about them, but say far less about binary stars. Models of planet formation about single stars will be reviewed briefly. Binary companions can disrupt the stability of planetary orbits. Nonetheless, stable planetary orbits with semimajor axes much smaller than the binary separation exist near the individual stars, and circumbinary orbits with semimajor axes several times larger than the stars' separation can also be long-lived. (In theory, stable triangular Lagrangian point orbits could also exist in binaries with mass ratios greater than 27.) However, orbital stability, while necessary, is not a sufficient condition for planets to exist. Binary companions may excite eccentricities and inclinations of planetesimals to such an extent that collisions lead to disruption rather than accumulation into planets. The effects of binary perturbations on planetesimal motions will be discussed, using interactions observed between moons and planetary rings as a guide.

Planets Around Main Sequence Stars

The first five planetary objects have now been found around Solar type stars. All have been found by monitoring Doppler shifts of FGKM main-sequence stars with a precision of about 10 m/s. The Doppler measurements yield orbital period, semimajor axis, eccentricity, and the quantity, M sini, where M is companion mass and i is orbital inclination. Within 5 AU, all five companions have M sini between 0.46 - 10 Jupiter Masses. No companions have been found having M sini between 10 - 40 Jupiter Masses, despite their ease of detection. Thus, the 5 companions represent a new population of extremely low-mass objects, having a discontinuous mass function from the ``brown dwarfs'' within 5 AU (none yet detected). Several prospective subclasses are emerging, having small orbits, high eccentricities, or circular orbits. One system shows evidence of a second low-mass companion, without a full period covered to date.

Pre-Main-Sequence Spectroscopic Binaries: Orbits and Environments

To date the only orbital elements for pre-main-sequence (PMS) spectroscopic binaries have been derived spectroscopically. One essential role of orbital elements is to derive stellar masses, and thereby to constrain theoretical stellar evolution models. To date precise masses have been derived for only two low-mass PMS stars. Other less direct (e.g., mass ratios and limits) or statistical constraints on PMS stellar masses are also available. In the first part of my talk I will review existing measures of PMS stellar masses, and the consequent contraints on PMS stellar evolutionary tracks.
In the second part of my talk I will turn to disks associated with short-period PMS binaries. Jensen will discuss the distribution of disk material Tuesday morning. I will focus on the issue of accretion flows in binary environments. Many short-period binaries are surrounded by circumstellar disks, in some cases with masses similar to or exceeding the minimum mass solar nebula. Several of these binaries show evidence for continued accretion of material at the stellar surfaces. Such accretion is a challenge to dynamical theory, since tidal and resonant torques driven by a binary tend to truncate the inner edge of a circumbinary disk and prevent further inward flow of circumbinary material. Without replenishment, any (small) circumstellar accretion disks would quickly exhaust themselves. Recently Artymowicz and Lubow have found a mechanism permitting pulsed accretion streams from circumstellar disks across a binary orbit onto circumstellar disks or stellar surfaces. At the same time we have found a short-period binary - DQ Tau - which both brightens and bluens at each periastron passage. As I will show, the similar phasing of the theoretical accretion flows and the observed brightenings is striking.

Low-Mass Secondaries and Planets: Mass and Eccentricity Distributions

We present a preliminary analysis of the distribution of secondary masses for spectroscopic binaries found in a radial-velocity survey carried out with the Digital Speedometers at the Harvard-Smithsonian Center for Astrophysics for stars in the Carney-Latham proper-motion sample. We find that the secondary mass distribution rises towards the lower end. It is not clear yet whether this rise continues below the stellar/substellar limit. Jupiter enable us to compare the orbital characteristics of this group with the corresponding characteristics of stellar secondaries in binaries. In particular, the eccentricity versus mass distribution of the two populations is very illuminating, and could possibly be used to distinguish between planets and low-mass secondaries.

Circumstellar disks in the Orion Nebula

We will present observations of the Orion Nebula made with the Hubble Space Telescope in which a number of stars in the Trapezium Cluster are surrounded by dark silhouettes seen projected against the bright background HII region. We find a variety of morphologies, all consistent with thin circumstellar disks spanning a range of diameters (50 to 1000 AU) and inclination angles (0 to > 80 degrees). The silhouette intensity profiles cannot be fit by standard disk models in which the surface density follows a radial power law with an exponent in the range -0.75 to -1.5. Rather, the data are best fit by opaque inner disks with exponential edges, and we discuss possible origins of this apparent truncation. Masses in the range 6 x 10²⁶ - 4 x 10³⁰ g (i.e. up to 0.002 M_Sol.) are determined for the disks by assuming that the faint light measured from them is background light transmitted through the disk. However, these are strict lower limits on the true disk masses, as most of this light can be accounted for by PSF blurring and scattering in the HST optical train: the present observations appear in fact to be consistent with completely opaque disks. Central stars are seen directly in five of the silhouettes, while the presence of a star is inferred in the sixth, where small reflection nebulae are seen above and below the plane of the near edge-on disk. Optical and near-infrared stellar photometry is consistent with young ( ~ 1 Myr) low-mass (0.3-1.5 M_Sol.) stars, with several showing evidence for excess near-infrared emission from the disk inner edge. While none of the stars in this small sample is seen to be binary (with separations in the range 100-1000 AU), we will also present preliminary results from a survey for binaries in a much larger sample of stars in the Trapezium Cluster, searching for differences in the binary fraction between those that do and those that do not show evidence for extended circumstellar disk material.

Searches for the Youngest Binaries

The expanding capabilities of millimeter and sub-millimeter wavelength arrays are making it possible to probe the earliest stages of star formation with resolutions as good as 0.3 arcseconds. These data are providing the best views of the earliest stages of binary formation since the surrounding opaque molecular cloud cores severely limit the observability of such systems at optical and near infrared wavelengths. To date, there are clear examples of binary systems with separations from 5 arcseconds down to 1 arcsecond (750 AU to 300 AU). There are also examples of apparent single star systems; although millimeter and submillimeter wavelengths only trace dust emission so a more evolved secondary component is always possible. Recent results for a number of systems will be reviewed and an attempt will be made to put these results into a statistical context.

Binary T~Tauri stars in the ROSAT Survey

With the spatially complete but flux-limited ROSAT All-Sky Survey (RASS), two-thirds of the weak-line T~Tauri stars (wTTS) and only one sixth of the classical T~Tauri stars (cTTS) have been detected in the neay-by Taurus-Auriga star forming region. With the ROSAT PSPC, one cannot resolve multiple TTS; however, recent (optical and IR) multiplicity searches among TTS show that most TTS are multiple. The RASS detection rate of known TTS binaries is significantly larger than for TTS that appear to be single. We conclude that binary TTS are overabundant among X-ray selected TTS. Similarly, TTS binaries among cTTS unresolved in optical observations are overabundant among H_alpha selected TTS. These biases have to be taken into account when estimating the binary frequency of cTTS and wTTS and also when comparing star forming regions, as the fraction of X-ray selected TTS among all TTS may be different from region to region. Additionally, only from RASS detections rates, we can estimate the number of cTTS/wTTS and cTTS/cTTS binaries among those TTS binaries, where the primary is known to be a cTTS and the secondary is of unknown nature. We predict binary fractions among RASS selected TTS and discuss these predictions in the light of most recent multiplicity studies among RASS discovered TTS both in star forming clouds and off the clouds.

Evolution of Inner Accretion Disks in Young Binary Systems

Large near IR color excesses and emission line spectra indicate the presence of optically thick circumstellar accretion disks and active boundary layers in young stars. We investigate whether this classic T Tauri (TT) behavior is found in one or both members of a binary system previously identified as a TT on the basis of the unresolved light. We consider the angularly resolved Br_gamma and, in one case, Na I, spectra of several close (1.3-2.6 '') TT systems. We also take into account the angularly resolved near IR (K-L) colors of a sample of additional young binaries with separations between 0.3 '' and 2.5 ''. We find that for all systems, both components show, or have shown in the history of their observation, evidence for classic T Tauri behavior. We demonstrate that this cannot be the result of random pairing of a population of single TT and weak-lined T Tauri (WT) stars. We speculate that the result that inner, AU-sized disks tend to survive for a similar length of time in both components of a close (0.3-2.6 '') binary suggests that a circumbinary envelope effectively regulates the common evolution of the inner disks.

Spectral appearance and diagnostics of extrasolar giant planets

The truly stunning discoveries of planets with radial velocity and astrometric surveys are motivating intense technological developments to meet the challenge of directly detecting an extrasolar giant planet (EGP). While the masses and orbital parameters of the new planets fall in unexpected ranges and are unpredicatable, models of the evolution and atmospheres of EGPs are valuable for guiding direct detection searches.
Jupiter has traditionaly been used as a benchmark for observability, but EGPs are now known to have a wide range of masses. Detailed evolution and simple flux estimates for EGPs show that young and massive planets can be significantly brighter than Jupiter and are the favored targets for direct detection. These models also indicate that the next generation of space- and ground-based telescopes should be sensitive enough to detect the emission of EGPs.
The spectral modeling of the first bona fide brown dwarf, Gl 229B, is illuminating. With T_eff ~ 950 K, it falls squarely in the transition regime between very late M-dwarfs and Jupiter, between stellar and planetary atmospheres. This is shown by its intermediate atmospheric composition and the absence of condensates near the photosphere. Modeling Gl 229B brings to light the shortcomings of current opacity databases which put strong limits on the accuracy of the computed spectra.
Preliminary synthetic spectra for several of the new planets indicate that absorption by CH₄ and H₂O causes very large departures from the blackbody flux. A judicious choice of bandpass however, can greatly favor detection. We find that the well-known 4-5 microns opacity window seen in Jupiter is shared by all objects up to Gl 299B. This appears to be a general diagnostic feature and we anticipate that it will be a good indicator of effective temperature.

Ground Based Interferometry at JPL

This paper summarizes the activities at the Palomar Testbed Interferometer and plans for the Keck Interferometer. The Palomar Interferometer PTI observed first fringes in July 95 and has operated in the dual star mode observing two stars simultaneously. We are now in the process of automating the operation of the interferometer in preparation for the narrow angle astrometry tests. We are also planning to make the instrument usable from 1.6 as well as 2.2um. Lastly, we are planning to install single mode fibers as spatial filters to improve the accuracy of the visibility measurements. The paper will end with the current status of planning/design activities for the Keck Interferometer.

The Binary Frequency in Open Clusters

It is plausible, though not necessarily proven, that most stars form in rich clusters. It is further accepted that even rich open clusters are transient phenomena, with lifetimes typically less than one Gyr. Plausibly, therefore, most field stars may have been born in rich clusters. Because dynamical processes act to disrupt, rather than create, binaries, the binary fraction as determined from open clusters can be used to place a lower limit on the fraction of stars that are born as binaries. I will review the attempts to determine the binary fraction in open clusters by photometric, spectroscopic, imaging and other techniques and I will try to assess whether there is evidence for significant variation of this fraction from cluster to cluster.

Evolution of Circumstellar Disks Due to the Gravity of Companions

We study the evolution of a circumstellar disk due to tidal interactions between the disk itself and an embedded protoplanet. In particular, the following two processes are analyzed in detail:
(1) Solutions for the propagation of density waves excited by the gravity of the protoplanet and their subsequent viscous damping are obtained within the context of the WKB approximation. The angular momentum carried by these waves and the torque exerted on the disk through the damping of the waves are obtained.
(2) The evolution of the disk arising from angular momentum transfer by the density waves is simultaneously computed by solving the hydrodynamical equations in an infinitesimally thin and non-self-gravitating disk with a Keplerian rotation law.
The propagation distance (or damping length) of the waves can be quite large in disks with low viscosity. Indeed, we find that with a viscosity parameter, alpha < 10^-3, m=2 waves can reach the inner edge of the disk, whereas for alpha < 10^-4, m ~ (r x Omega / c)_p ~ 20 waves can reach the inner edge as well. In disks with such small values for the viscosity, the gap size is determined by the damping length of the waves. The gap size becomes wider as the disk viscosity is decreased. When the viscosity becomes small enough to allow the waves to propagate to the inner region near the stellar surface, the removal of the inner disk ensues. We find that a Jupiter mass protoplanet facilitates the removal of the inner disk if alpha < 3 x 10^-4, and that the time scale for this depletion is 10⁵-10⁶ years. The inner disk depletion will probably be detected from the future radio observation of circumstellar disks with high angular resolution of 10 milli-arcseconds. We also determine the conditions for the gap formation in terms of relations between the masses of the protoplanets and the properties of the circumstellar disks.

A search technique for planets in nearby binary systems using a ground-based interferometer

A search for Jovian-type planets in 100 nearby binary stars could be carried out with the existing ground-based Infrared-Optical Telescope Array (IOTA) interferometer. We would study binaries with sufficiently great separation (25-50 AU; typical separation around 0.4 arcsec) that such a planet could be in a stable orbit about one member of the pair. The method is to measure the angular separation of stars in each binary, with a single-measurement accuracy sufficient to detect the amplitude of a Uranus orbiting one of the stars. The technique is based on an auxiliary device, the Pupil-Splitting Interferometer (PSI), which substantially reduces systematic and random errors by converting a measurement of angular separation into a measurement of the differential optical delay between the two components of the binary. The program would be relatively economical, and could begin soon. In a preliminary experiment, we recently observed a few of the binaries on our list, including the full range of magnitudes, using the IOTA interferometer, but without the auxiliary (PSI) interferometer, which has yet to be built; initial results will be shown.

ABSTRACTS FOR THE POSTER PAPERS

Giant Gaseous Protoplanet Formation?

Radiative hydrodynamical calculations of the thermal structure of an axisymmetric protoplanetary disk with a mass of ~ 0.14 M_sol orbiting a solar-mass star predict that the outer disk may be cool enough (~ 100 K) to become gravitationally unstable. This possibility has been investigated with a fully three dimensional hydrodynamics code. Growth of nonaxisymmetry occurs within a few rotation periods of the outer disk (P_o 30 yrs); the nonaxisymmetry is large enough to result in disk evolution through gravitational torques within 10⁴ yrs. After about 10 P_o, the dominant m = 1 and m = 2 modes saturate at an amplitude greater than 1 - by this time, a 6-Jupiter-mass clump of gas has formed around 8 AU, accompanied by a 0.4-Jupiter-mass clump at the same orbital radius, but 180^o away in azimuth. The clumps are gravitationally bound and so should form giant planets. The hot inner disk remains nearly axisymmetric throughout, suggesting a ``best of both worlds'' scenario where formation of terrestrial planets occurs through collisional accumulation in the hot inner nebula, while rapid formation of giant gaseous protoplanets occurs in the cool outer nebula through gravitational instability of the disk. However, the gravitational instability appears to depend on the thermodynamics assumed for the azimuthal density variations - if these variations occur adiabatically with an effective gamma = 7/5, the instability does not appear to lead to clump formation, whereas for gamma = 1 (isothermal) the instability is robust. Even for isothermal azimuthal variations, the instability is damped when the outer disk is hotter than about 150 K. Realistic values of gamma fall between the two extremes of gamma = 7/5 and gamma = 1; further 3D models with intermediate values of gamma are in progress to determine if this instability is likely to lead to giant planet formation.

Disk Clearing in a Binary Environment

Tidal force from a companion may suppress the formation of a protoplanatary disk or disrupt an existing one. On the other hand, resonance excitation by the companion may be an efficient way to build up the density contrast in the disk, hence enhancing the accretion activity, and facilitate the condensation of protoplanets. We model a young star's SED considering continuum emission from a geometrically thin, optically thick disk coplanar to the binary orbit.
In general the disks surrounding single stars are more active in accretion, with a radial temperature distributions shallower than passive, reprocessing or steady-state accretion disks (q=0.75, T(r) ~ r^-q). For 0.1-10~Myr old singles, 0.5 < q < 0.7. In contrast, binaries have a wider range of variation 0.5 < q < 1.2. Specifically, spectroscopic binaries ( 0.5 < q < 0.9) are similar to single stars in terms of disk accretion, except that in such a system a central hole is cleared up and the accretion occurs in the circumbinary disk which contributes the infrared emission. Our modeling of the SB V4046 Sgr shows q ~ 0.5, a disk inner radius of 0.06-0.09 AU, outer radius of 20-30 AU and an accretion rate 1-5 10^-8M_sun/yr. The size of the central hole (2-3 times of the binary separation) is consistent with dynamical clearing by the binary. Incidentally our near-IR images of another SB AK~Sco show a source 20" to the SW. Therefore an extended IR component to accommodate the unusual (bulging) near-IR excess is rendered unnecessary. For binaries with separation 1-100~AU, 0.6 < q < 1.2 . A hole in the circumbinary disk can still be evacuated, but now the accretion occurs in the circumstellar disk(s) which can not be replenished. Accretion thus diminishes as the available material is exhausted ( ~ 1 Myr). More massive stars, due to higher exhausting rates, will cease accretion activity at an earlier time in evolution. For wide systems (100 AU) the situation is again similar to that for single stars, namely there is plenty of matter for circumstellar accretion (over 10 Myr), but still not replenishable. Their q values ( 0.5 < q < 0.7) therefore increase (i.e., accretion subsiding) as the stars age, and as the stellar mass decreases.
In many wide (~ 1") binaries the primaries (the more massive ones) tend to be older, hotter and more reddened in the J-H, H-K colors. If they are physical pairs, this implies that the companions are formed noncoevally, perhaps out of the disks, and now located in regions relatively void of cloud material, whereas the circumprimary disks remain prominent.

The TEP Project - a Search for Transits of Extrasolar Planets

The TEP (Transits of Extrasolar Planets) collaboration was formed in 1994 with the goal of a photometric detection of transits of extrasolar planets around the eclipsing binary CM Draconis. This star is the lowest mass eclipsing binary known (dM3/dM4). It is also relatively close (17 pc), has a period of 1.3 days and its orbital plane is nearly within our line of sight (i=89.7^o). These conditions give a unique opportunity to determine the existence, or non-existence, of planets around this binary by photometric means with a high degree of certainty. Planetary orbits, if present, will be within the orbital plane of the binary components, and - due to the small size of the components- create photometrically detectable transits. The transit of a Jupiter sized planet will cause a brightness drop of 1%, and an earth sized planet one of 0.1%, which is detectable with subnoise detection algorithms. This method is currently the only feasible one to detect terrestrial planets around normal stars. The low mass and temperature of the binary components leads to the expectation, that comparable planets will form much closer to the central stars than in our Solar System, and will have orbits of a few weeks to months. An observing campaign gathering several months of CM Dra lightcurves will lead to a high probability to detect a planetary transit, even when the data are taken non-continuously. The TEP collaboration has been observing CM Dra from several 1m-class telescopes in the last two years and observations at larger telescopes are scheduled for summer 1996. So far, data covering about 700h have been obtained and most of them reduced. Several signatures in the lightcurves have been found that are not incompatible with planetary transits. After conclusion of the observations in spring-summer 1996, we should be able to set limits with reasonable confidence (> 80%) on the existence or non-existence of planets around CM Dra. As second method that is being followed up consists of measuring the time-shifts of eclipse minima around CM Dra and other eclipsing binaries. The time-shift is caused by the motion of the binary system around the barycenter if a third object is present. For planets with Jupiter masses and orbits, the shifts will be on the order of several seconds over a half-period of the planet.

Multiplicity of TTS in Southern Star Forming Regions

High resolution studies of young stars in the star forming regions of Taurus and Ophiuchus have revealed a large population of multiple star systems. To test how ubiquitous this earlier result is for other star forming regions, we have carried out a K[2.2 mu m] band multiplicity survey of T Tauri stars located in the dark cloud complexes Chameleon, Lupus, and Corona Australis. This survey, which was conducted with both speckle and direct imaging techniques, covers a binary star separation range of 0."1 to 12" (14 - 1680 AU) and identifies 24 companion stars of which 13 are new detections. Over the separation range covered, the T Tauri binary star frequency is estimated to be 0.52 +/- 11, which is in agrees with earlier measurements for other regions (0.58 +/- 0.08 for Tau & 0.50 +/- 0.12 for Oph). Although the total binary star fractions for these regions are similar, they appear to differ in their distributions of separation as well as flux ratios. A comparison of the direct imaging portion of this survey with Reipurth & Zinnecker's (1993) optical multiplicity study also reveals that 8% of the overlap sample have infrared companions (i.e., companions which are too red to be detected at optical wavelengths).

High Spatial Resolution Imaging of Pre-Main-Sequence Binary Stars: Resolving the Relationship Between Disks and Close Companions

Although the standard model for star formation is based on single stars, we now know that most T Tauri stars are in multiple systems with the majority having separations less than 100AU, the size typically inferred for a circumstellar disk. These close companion stars have the potential to significanly alter their circumstellar environments. Furthermore with angular separations less than 1", they are unresolved in most types of observations. Resolving the individual components over a wide range of wavelengths would permit unbiased estimates of their stellar and circumstellar properties. This is useful for constraining possible binary star formation mechanisms as well as the potential for planetary formation in these close multiple systems. We have therefore carried out a high spatial resolution imaging study of six T Tauri binary stars with separations ranging from 10 to 50 AU using the Hubble Space Telescope from 0.3-0.9 microns and speckle imaging from 1-2 microns These complementary observations show that the secondary stars can contribute significantly to the unresolved spectral energy distributions and thus bias the previously inferred stellar ages and circumstellar properties. For all the systems studied here, the individual components are consistent with being coeval, suggesting they formed from the same cloud core. For the four classical T Tauri stars all components show signatures of an inner circumstellar disk.

Star-Disc Collisions - Bad News for Planets?

We investigate the response of a circumstellar accretion disc to the fly-by of a perturbing mass on a parabolic orbit. Penetrating encounters are highly destructive to discs, removing as much as 70 percent of the material in one encounter. The effect of all encounters, particularly penetrating ones, is to steepen the density profile, increasing the surface density at small radii whilst extensively stripping the disc beyond periastron. In all geometries studied, we find that for penetrating encounters the surface density profiles outside periastron are well fit by exponentials. These agree well with recent observations of truncated discs in the Orion Nebula. Repeated passages at the same periastron separation have little effect on the remaining material - it is the first encounter that dominates both mass redistribution and energy and angular momentum transfer. In almost all close encounters the energy and angular momentum transferred between orbit and disc material are dominated by disc material becoming unbound from the system. The magnitude of the energy transfer calculated in our simulations is greater than that of the binding energy of material exterior to periastron by a factor of between two and five. The destructive nature of the encounters indicates that a non-linear treatment is essential.

Rotation and Variability of Stars in the Orion Nebula Cluster

For the past five years we have monitored pre-main sequence variables in the Orion Nebula Cluster with the 0.6 m Perkin telescope and CCD at Van Vleck Observatory on the campus of Wesleyan University. Rotation periods with false alarm probabilities of less than 1% have been determined for more than 130 stars, many at multiple epochs. In no case do we find a significant change in the period from one epoch to another, except for a single example of period doubling. On the other hand, periodicity is only rarely seen at all epochs. When non-periodic, the stars generally maintain their same average brightness level and variability range, suggesting that spots continue to be present but not in sufficiently stable patterns to produce a detectable period as sampled by us. The suggestion by Choi & Herbst (1996) that there is a variation in rotational properties within the ONC is explored with our larger data sample. Results of a search for eclipsing binaries are discussed. No definite examples have been found, but candidates for spectroscopy are proposed.

The Astrodynamics of New Planetary Systems

Evaluation of possible orbits for extra-solar planets stable for life is analogous to the astrodynamics analysis problem of determining a satellite mission orbit. For the extra-solar problem, the analyst is concerned with the gravity and radiation fields of the primary and, in many cases, perturbing effects of a secondary body, whether a binary star, a giant (jovian mass) planet or brown dwarf. Beginning in 1985, the author analyzed planetary stability conditions for nearby binary systems (Alpha Centauri, Sirius, Procyon...) as a Restricted Elliptic Three Body Problem (RE3BP), characterized by mean stellar separations 20-30 A.U. and system eccentricities 0.3 - 0.55. Hypothetical planets in orbits thermodynamically analogous (Teff = 400 K) to Earth's were initialized and tracked for thousands of years, along with planets placed in other temperature contours(350, 450 K). Results indicate that secondary stars perturb nominally circular planetary orbits at stellar pericentron passage, causing planetary eccentricity and angular momentum to cycle over thousands of years as argument of planetary periastron rotates 360^o in the celestial sphere. With recent interferometric and other observations having detected massive planetary or brown dwarf partners to stars at radii where terrestrial planets were preferred (47 Ursae Majoris, 70 Virginis, 59 Pegasi, Gliese 239...), RE3BP studies now include the new cases and better characterize binary environments. With accelerations proportional to M/R2, the effects for the first two "planetary" cases, despite less mass, were similar to the binary star examples due to closer proximity. Comparisons of several binary systems show that regions where planetary orbits are unstable or bodies are ejected from the system, vary in control volume temperatures, affecting the available chemical constituents of protoplanetary material. Regions of stable orbits also appear to have many quantum-like properties from which phase diagrams can be derived. Thus, terrestrial planets, if extant near binaries, would experience complex climate cycles influenced by secondary bodies and different circumstances for formation, whether processes are accelerated or inhibited.

Infrared Colors of Pre-Main Sequence Binary Stars

Critical to any discussion of the formation of planets in the circumstellar environment of binary stars is an evaluation of the evolutionary status of the stars themselves. This paper presents the preliminary results of new high-resolution near-infrared imaging observations of a set of recently-discovered pre-main sequence binary systems in the Ophiuchus star-forming region. The goal is to begin to determine the masses and ages of the components and test their consistency with a coeval formation mechanism, and to measure the location and properties of any warm circumstellar disk material.

The Response of an Accretion Disc to the Tide of an Orbiting Companion

Duplicity in a X-ray selected sample of young stars in the Upper Scorpius region

Fred Walter (AJ 107, 692, 1994) has studied a sample of 28 X-ray selected young low-mass stars in the Upper Scorpius OB association. He found them younger than the associated B stars and coeval within +/- 1 Myr. One possible explanation of these observations could be the sudden removal of gas and dust from the region by a supernova explosion. Conceivably this could hinder the formation of binaries.
We searched the stars in this sample for companions by means of near-infrared speckle interferometry at 2.2 microns. We find no obvious decrease in duplicity. This is at odds with the imaging 1 micron survey by Brandner et al. (A&A 307, 121, 1996), who find for this region the same comparatively low degree of duplicity as known for solar type main-sequence stars.
Points to be discussed are the significance and meaning of the results and of the apparent discrepancy.

The companions to the nearby M dwarf LHS 1070 - cool and moving

Infrared speckle observations had shown that the nearby M5.5 dwarf LHS 1070 has two companions at ~ 1 '', separated among them by ~ 0.4''. Based on the absolute K magnitudes of 10.5 mag and 10.8 mag for the components, we estimated the masses of components B and C to be close to the hydrogen burning mass limit of 0.08 M_sol (Leinert et al., A&A Letters 291, L47, 1994).
We now present preliminary evaluations of BVRIJHK photometry of the components and of optical spectroscopy between 400 nm and 850 nm. We use both, qualitative arguments and comparison to model atmospheres, to give first conclusions regarding effective temperature and composition.
We use the preliminary results on the relative motion between components B and C and try to estimate the mass sum of this close pair within the triple system.

BIMA High Resolution Observations of L1551-IRS5

BIMA extended array configuration observations of L1551-IRS5 at lambda=2.7 mm have been obtained. In this configuration the longest baseline was 480 k lambda, providing a synthesized beam of 0.69'' x 0.32''. The millimeter emission is not concentrated in the center of the lambda=2 cm extended emission seen by Rodriguez et al. (1986).
We detect emission from two compact sources, coinciding with the centimeter sources, separated by 0.35''. The flux ratio for our two sources is similar to the flux ratio of the sources at centimeter wavelength.
With these observations we are not able to determine if the source L1551-IRS5 is a binary system or an isolated star.

Star/Planet Tidal Interactions

Planets like 51 Peg (4.2d orbit period) are subject to potentially powerful tidal forces from the companion star. The orbit of 51 Peg is now formally unstable to decay, but on a timescale greater than its age. It could survive even for an orbit period of 10 h. In the past, the star was bigger and survival at such periods is problematic, thereby constraining the star's early spin history. We propose a new spin-down mechanism for the planet. Due to heating from the star, the planet's outer layers are convectively stable.
At the boundary between the convective interior and the outer layers, the planet is subject to a powerful resonant torque which despins the planet to near synchronism on timescales less than its age.

Aperture Synthesis Observations of Dust and Gas Around Intermediate-Mass Pre-MS Stars

Various on-going imaging surveys of protoplanetary disks are concentrating upon T~Tauri systems with stars of spectral types~M and~K and corresponding stellar masses in the range 0.25 (M_*M_sol)² (e.g. Handa et al. 1995; Koerner & Sargent 1995; Dutrey et al. 1996).
A new and complementary program of observations with the Owens Valley mm-wave array is now extending the search for disks to sources of earlier spectral type and therefore greater stellar mass. By moving across the pre-MS region of the H-R diagram in this way, we aim to investigate the possible broad dependence upon M_* of not only the bulk properties of disks (principally their masses and the survival times of gas and dust components) but also the potential for the creation of planetary systems.
In the past year we have used the Owens Valley array to measure dust continuum emission and CO spectral lines at lambda=3 mm, and we have examined the dust and gas located within several hundred AU of 7~young stars with spectral types~F and~A and stellar masses of around 2-4 M_sol (Mannings & Sargent, in preparation). All of the sources are detected in continuum, and four are detected in CO line emission. The emitting regions are centered on the positions of the stars.
Substantial masses of gas and dust exist near to each star. We estimate dust+gas masses in the approximate range 0.01-0.1 M_sol. For such masses, the V-band extinctions through, for example, spherical envelopes surrounding the stars would be severe (A_V ~ 50-1000 magnitudes). Since six of our sources are easily visible at optical wavelengths, the implication is that the majority of the circumstellar emitting regions we have observed are both flattened and inclined.
These observations are very encouraging, and they suggest that possible protoplanetary disks around young stars might not be restricted only to solar-mass T~Tauri systems. We are currently mapping further sources and will observe young stars of spectral type~B, extending the sampled range of stellar masses up to 10 M_sol.

Dutrey et al. 1996, A&A, in press
Handa, T., et al. 1995, ApJ, 449, 894
Koerner, D. W., Sargent, A. I. 1995, AJ, 109, 2138

An Infrared Search for Brown Dwarf Companions to Nearby Stars and White Dwarfs

As part of an ongoing search for brown dwarf companions, we have concentrated on two populations: nearby (d<70 pc), young main sequence stars and nearby (d< 50pc) white dwarfs. We are observing the vicinity of about 100 young (< few x 10⁸ yrs) late type main sequence stars, in J and K, using a coronagraph to block the light from the primary. In the case of the few hundred white dwarf targets, a coronagraph is not needed since the infrared fluxes are comparable to those expected for brown dwarfs. Candidate objects are re-observed at a later epoch to determine if they share the proper motion of the primary. We employ the Steward Bok 90 '', the IRTF, and the Lick 3m telescopes to reach limiting magnitudes of ~ 19.5 in J and ~ 18.3 in K. Other than GD 165B, no strong brown dwarf candidates have yet been identified.

Spiral Instabilities and Clumping in Massive Accretion Disks

We present a series of hydrodynamic simulations modeling the growth of instabilities in massive disks around single stars using a Smooth Particle Hydrodynamics code. These simulations show that spiral arms of order m < 3 dominate in massive disks, while filamentary spiral structure of high order m>>3 is seen at low disk mass. The boundary between these regimes is seen at the predicted `maximum solar nebula' mass of Shu, Tremaine, Adams and Ruden (1990). The simulations end in spiral arm collapse into one or more clumps whose mass comprises some 1% of the mass of the system. No final state of the system is determined. An isothermal (gamma=1) equation of state is assumed and we enforce an r^-1/2 temperature law at all times. This assumption implies very strong cooling in the disk and limits the physical interpretation that can be placed upon clump formation. Our results therefore represent a bracket beyond which processes become progressively less physically realistic.

The Multiplicity of Stars in the Hyades: Tracking the Evolution of the Binary Star Frequency

To address the intriguing discrepancy of binary star fractions between T Tauri stars and Solar Neighborhood G-dwarfs, we have conducted a speckle imaging survey of the Hyades, a cluster with an intermediate age (~ 6 x 10⁸ years). This survey of 162 stars can detect companions with a K magnitude difference of 4.0, which corresponds to a mass ratio of 0.3, and separations of 0.10 to 1.07 arcseconds, or 5 to 53 astronomical units. Over this range of separations, 26 binary stars are detected, resulting in an observed binary star fraction of 0.16. By assuming that the Hyades companion mass distribution is similar to the Solar Neighborhood mass function, we can correct for unseen companions and derive a binary star fraction of 0.25. This is consistent with a secular evolution of the binary star fraction from the T Tauri stars (0.40 at a few million years) to the nearby G-dwarfs (0.14 at a ~ 9 billion years), which would imply that stars observed as single today were not necessarily formed as single stars. This is particularly important if these early companion stars reduce the potential for planetary formation before they are lost. This work is currently being extended to other young clusters, such as the Pleiades and alpha Per.

The PMS Double-Lined Spectroscopic Binary NTT 155913-2233

Using high resolution, near infrared spectroscopy, we detected the secondary component of the optical spectroscopic binary in the pre-main-sequence system NTT 155913-2233, thereby converting the single-lined spectroscopic binary into a double-lined one. We identify approximate spectral types of K5 and M5 for the stars in the double-lined spectroscopic binary (SB2), and K5 for the tertiary component at 0.29''. We find a mass ratio of 2.0+/-0.4 for the SB2. From the mass function determined by Mathieu (1994) we obtain lower limits for the primary and secondary masses of 1.1+/-0.2 and 0.6+/-0.1, respectively. Using fluxes determined from our model fit to the data, together with the spectral types and the mass ratio, we find that the NTT 155913-2233 SB2 is coeval, within error, when plotted on the Swenson et al. PMS evolutionary tracks, but not on the D'Antona & Mazzitelli tracks.

Binaries among Herbig Ae/Be stars

We have studied the circumstellar environment of 31 Herbig Ae/Be and related stars in the near infrared by means of speckle-interferometry. For the brighter objects we reach or approximate diffraction-limited resolution of ~ 0.1'' (typically 100 AU at the object). Eleven objects have companions, five of which constitute sub arcsec binaries, mostly detected by us. This study can be considered an extension to higher masses, of the recent surveys for binarity among T Tauri stars. In the case of the Herbig Ae/Be stars the task is more difficult, because of the smaller number of sources available and their larger average distance. Although our sample is small and neither homogeneous nor complete, we try to draw some preliminary conclusions. In particular, it appears that also for Herbig Ae/Be stars, as is now evident for T Tauri stars, there is a high incidence of binaries.

Stability of accreting triples

We present a preliminary analysis of the stability of young triple systems undergoing accretion. Recent observations have shown a large number of triple systems amongst pre-main sequence stars. Such systems are unstable if the binary separation is an appreciable fraction of the binary-tertiary separation. We show that such systems can evolve towards stability due to accretion onto the system, dependent on the angular momentum of the infalling matter. If the angular momentum is low, then the separation of the binary will decrease faster than that of the binary-tertiary. This will move the system towards stability. If the angular momentum of the infall matter is high, then the tertiary will accrete the mass and angular momentum of the infalling matter, and thus increase its separation from the binary. We discuss the implications of this analysis on the relative population of triple systems.

Notes on the stability of planets in the Alpha Centauri system

This poster investigates the long-term stability of small bodies in the Alpha Centauri system. The system is a triple star with two of the stars forming a close binary (23 AU) and the third orbiting the pair at a much greater distance (12,000 AU). We seek regions of phase space where test particles (planets) could remain for times on the order of the ages of the stars (10⁹ - 10¹⁰ years). The goal is to determine if we should expect to find remnant material at observable distances from the stars. As Alpha Cen is the closest star system to the Sun, it is a prime place to prospect for planets. Stability considerations constrain the locations where planets are likely to exist, increasing the efficiency of telescopic searches.

X-ray detected and other PMS stars in Orion OB1.

Results are presented from recent photometric observations of x-ray sources in the Orion OB1a and OB1b star formation regions. Since the early 1980s, x-rays have been found to be the premier method of identifying pre-main sequence stars, especially in highly confused regions. The optical properties of 105 x-ray sources are discussed in detail. Based on optical colors, we find that the x-ray sources in both regions lie along a well defined locus above the main sequence. Infrared and spectral data confirm that 90% of the x-ray sources are PMS and about 10% of the PMS stars may have disks. We find that the x-ray data are 80% complete relative to the PMS population between V magnitudes 12 and 16. The completeness study also reveals a population of PMS stars to which the x-ray surveys were insensitive. These PMS stars range in mass down to near the brown dwarf limit. This photometric method, augmented by spectra, may supplant x-rays as the premier method for identification of PMS sources in regions of star formation. It may also provide a method for observing stars at the low mass end of the initial mass function.