Astron. Astrophys. 336, 411-424 (1998)

1. Introduction

An important problem in astrophysics concerns the nature of the dark matter in galactic halos, whose presence is implied by the observed flat rotation curves in spiral galaxies. Microlensing (Paczynski 1986) allows the detection of MACHOs in the mass range (De Rújula, Jetzer, Massó 1992) in the halo, disk or bulge of our galaxy. Till now, more than 15 microlensing events have been found towards the Large Magellanic Cloud (LMC) (Alcock, Akerlof, Allsman et al. 1993, Alcock, Allsman, Alves et al., 1997b, Auburg, Bareyre, Bréhin et al 1993), one event towards the Small Magellanic Cloud (SMC) (Alcock, Allsman, Alves et al. 1997c, Palanque-Delabrouille, Afonso, Albert et al. 1997, Udalski, Kubiak & Szymanski 1997) and about 200 events towards the galactic bulge (Alcock, Allsman, Alves et al. 1997a, Udalski, Szymanski, Stanek et al. 1994, Alard & Guibert 1997).

However, in spite of the many events, several questions are still open, in particular on the mass and the location of the lenses. In fact, from the duration of a single microlensing event, one cannot infer directly the mass of the lens, since its distance and transverse velocity are generally not known. To break this degeneracy it has been proposed to perform parallax measurements (Gould 1997), which however require the use of space satellites.

Globular clusters could be in many respect very useful to solve some of these problems. In fact, microlensing searches using globular clusters as targets could probe different lines of sight in addition to the ones towards the LMC or the SMC, this way allowing to better determine the spatial distribution of the MACHOs (Jetzer 1991). Since in globular clusters much less stars than compared to the LMC or SMC can be used as targets, one would have to monitor many globular clusters in order to get some microlensing events. Gyuk & Holder (1997) and Rhoads & Malhotra (1997) have studied this possibility and shown that this way interesting galactic structure information can be extracted allowing to distinguish between different halo models.

Another possibility is to search for microlensing of background stars by MACHOs located in foreground globular clusters. Such an observation can in addition give important information on the total mass of globular clusters. It has been argued recently that a large fraction of their mass (around 50%) is dark, which could be in the form of brown dwarfs, low-mass stars or white dwarfs (Heggie et al. 1996, Taillet, Salati & Longaretti 1995, 1996). Moreover, one expects that the heavy stars tend to sink towards the cluster cores, whereas the light objects populate the outskirts. Hence, the dark component of the cluster is not similarly concentrated towards the center as the bright stars which eases observation.

The idea -as originally proposed by Pacznski (1994)- is to monitor globular clusters like 47 Tuc or M22 in front of the rich background of either the SMC or the galactic bulge. In this case, when the lens belongs to the cluster population, its distance and velocity are roughly known. The velocity is defined by the dispersion velocity of the cluster stars together with the overall transverse velocity of the cluster as a whole. Knowing approximately the distance and the velocity of the lens would allow to extract from a microlensing event the mass of the lens with an accuracy of 30%.

Due to these reasons, it is important to study in more detail microlensing by globular clusters either using their stars as sources, or the dark matter contained in them as lenses for more distant stars (Wandeler 1995). In this paper we discuss both aspects in detail. Although the mass distribution of the luminous part of the cluster, as inferred from the observation of the distribution of the red giant population, agrees well with a King model, we do not consider this to be representative for the population of light objects. Taillet, Longaretti & Salati (1995, 1996) have shown that in an isolated globular cluster thermalisation between the different populations does occur. However, globular clusters -especially the ones towards the bulge- tidally interact with the surrounding material which might counteract thermalisation, hence the consideration of alternative mass distributions should be taken into account and variations of the microlensing event rate due to it might give some hints at the dynamical history of the cluster.

We also analyse the microlensing events towards the galactic bulge, which are close to the three globular clusters NGC 6522, NGC 6528 and NGC 6540. These clusters lie within the observation fields of the MACHO and OGLE teams. We find evidence that some microlensing events are indeed due to MACHOs located in the globular clusters, suggesting therefore that these clusters contain a significant amount of dark matter.

The paper is organized as follows: in Sect. 2 we introduce briefly the basics of microlensing. In Sect. 3 we discuss as an example the globular cluster 47 Tuc, which will then be used to estimate in a very conservative way the optical depth and the lensing rate for other clusters as well. In Sect. 4 we present microlensing using globular clusters towards the galactic bulge. In particular, we analyse the events as reported by the MACHO and the OGLE collaboration lying within a distance of pc around the centers of NGC 6522, NGC 6528 and NGC 6540. In Sect. 5 we conclude with a short summary of our results.

© European Southern Observatory (ESO) 1998

Online publication: July 20, 1998
helpdesk.link@springer.de