Previous
abstract 
Next
abstract
We are in the process of building a computational machine to model dense stellar clusters, as at the centres of galaxies, using the orbit-averaged Fokker-Planck equation in two-dimensional action-angle space. The model expands the work of Quinlan \& Shapiro (1990 ApJ 356, 483) to include rotation and associated effects: ellipticity and angular momentum. A non-axisymmetric component of the potential (i.e. the formation of a stellar bar) is invoked to increase the efficiency of angular momentum transport. Following Q\&S we plan to include of order 10 distribution functions, each representing a stellar population a factor of two more massive than the next; in addition there will be a separate, non-collisional bar-star distribution function; the other stars will interact with the bulk potential of these bar stars. The goal is to verify that angular momentum can be removed from the core of the cluster quickly enough so that the constraint of rotational support no longer prevents the formation of a massive $\sim1000M_\odot$ object at the centre. This would be the seed black hole for the formation of an active galactic nucleus.
