On the Reported Death of the MACHO Era

Although convincing evidence for the existence of dark matter has been around for over 40 years, its nature remains a mystery. If MAssive Compact Halo Objects (MACHOs) constitute a significant fraction of the dark matter budget, then a combination of observational and theoretical arguments constrain the properties of viable MACHO candidates to well-defined regions of parameter space. Microlensing experiments (eg Wyrzykowski et al. 2008;Tisserand et al. 2007) have, for instance, ruled out MACHOs with masses in the range 10 -7 -30 M as major constituents of the Milky Way's dark matter halo, thereby excluding dark matter candidates such as halo brown dwarfs or solar-mass black holes. In addition to microlensing, constraints from a number of indirect arguments such as the observed velocity dispersion in the disk (Lacey & Ostriker 1985), evaporation of low mass gas clumps ("snowballs") (Rujula et al. 1992), further reduce the parameter space available to baryonic Galactic MACHOs to ≈ 30 -10 6 M .

A recent analysis of the distribution of wide halo binaries in Yoo et al. (2004;hereafter Yoo04), failed to detect a clear signature of the disrupting effect of MACHOs on the widest, and hence most weakly bound, binaries. Consequently, the study appeared almost to close the door entirely on the remaining region of viable MACHO parameter space, leaving only a small window between 30 M and 43 M . A look, though, at Fig. 5 in Yoo04 suggests that their results depend critically on the validity of the two widest binaries in the observed wide halo binary sample from Chaname & Gould (2004;hereafter CG04).

In this letter, we present radial velocity measurements of the stars in each of these candidate binaries along with two other largeseparation halo binary candidates from CG04. Our radial velocities imply that three of these candidates are genuine binaries and we thus demonstrate directly that halo binaries with projected separations of ≈ 1 pc exist. However, our measurements also reveal that the second widest binary in CG04 is actually a spurious interloper. We update the constraints on MACHOs arising from these measurements. The removal of the spurious pair from the analysis eases the constraints significantly with the upper limit on MACHO mass increasing by an order of magnitude. In addition, the Galactic orbit we obtain for the widest binary raises questions on the validity of using this object in the analysis carried out by Yoo04. Its omission would re-open the region of parameter space closed in Yoo04. Furthermore, we also point out that, if the initial logarithmic slope of the binary separation function is set to -1, a choice with some theoretical foundation, then an un-evolved distribution is ruled out by the observations. The outline of this letter is as follows. In Section 2 we present our spectroscopic data for the candidate binaries, and in Section 2.2 we derive radial velocities for the pairs and use these to determine which systems are genuine binaries. In Section 3, we re-visit the constraints on the MACHO content of the Milky Way halo based on our new data. Section 4 summarises our conclusions.

The search for wide halo binaries is still in its infancy as a number of difficult observational challenges need to be overcome. First, halo stars are rare, constituting less than 0.2% of local stars (Helmi 2008). Second is the problem of distinguishing wide binary stars in samples of halo stars from mere chance associations. To date there has been one keynote study (CG04), which detected a large number of local (sample median distance is 240 pc), high probability candidate wide halo binaries (namely 116). The angular separation function of these binaries followed a power law distribution out to angular separations corresponding to ≈1 pc which was the detection limit of their survey. The candidate binaries in the CG04 sample were chosen from the revised New Luyten Two-Tenths Catalog (NLTT) of high proper motion stars, (Gould & Salim 2003;Salim & Gould 2003). Candidate halo pairs were required to satisfy proper motion consistency tests and to lie along what are essentially isochrones in the halo region of the reduced proper motion (RPM) diagram. As the angular separation, ∆θ, between the members in a candidate binary increases, the probability that the candidate pair is merely a random association increases roughly in proportion to (∆θ) 2 . CG04 argue that their halo binary sample is unlikely to be contaminated out to ∆θ = 900 . While the CG04 candidate binaries have angular separations smaller than this value, their importance for placing restrictions on viable MACHO candidates for dark matter demands that they be subject to further tests, in particular the candidate binaries with the widest angular separations which run the largest risk of misidentification.

One useful test to further explore the nature of these objects is to measure their radial velocities: both members of the candidate binaries should have essenti

…(Full text truncated)…

This content is AI-processed based on ArXiv data.