Space debris removal is very important for ensuring space safety and further space exploration. However, most of the debris is tumbling in space and its corresponding large mass and inertia result in non-negligible momentum. During the detumbling manipulation of large-momentum space debris, there will be large base disturbance force which makes the control of the floating base of space robots much more challenging. In this paper, we propose a trajectory optimization framework for stabilizing a tumbling object using a dual-arm space robot. In this proposed trajectory optimization framework, only the object dynamics is used to minimize the detumbling time and base disturbance during the detumbling manipulation, while the optimal dual-arm motion is generated from the object motion by considering the closed-chain constraint. The objective function of trajectory optimization is defined as weighted detumbling time and base disturbance force to save energy of the dual-arm space robot for further manipulation. In order to verify the proposed method, a typical simulation of dual-arm space robot stabilizing a tumbling target is carried out.