Classical mechanics and Time Dependent Hartree-Fock (TDHF) calculations of heavy ions collisions are performed to study the rotation of a deformed nucleus in the Coulomb field of its partner. This reorientation is shown to be independent on charges and relative energy of the partners. It only depends upon the deformations and inertias. TDHF calculations predict an increase by 30% of the induced rotation due to quantum effects while the nuclear contribution seems negligible. This reorientation modifies strongly the fusion cross-section around the barrier for light deformed nuclei on heavy collision partners. For such nuclei a hindrance of the sub-barrier fusion is predicted.Tunneling, the slow "quantum leak" through a classical barrier, is an intriguing phenomenon in nature. In 1928, Gamow discovered this effect looking for an explanation of the α radioactivity [1]. However, the tunneling of complex systems remains to be understood. As in the Gamow times nuclear physics is providing one of the most challenging field to understand tunneling. In particular, fusion cross sections involving massive nuclei around the Coulomb barrier can be orders of magnitude over one dimensional quantum tunneling predictions. Couplings between the internal degrees of freedom and the relative motion deeply modifies tunneling [2]. Neutron transfer, excitation of low-lying vibrational and rotational states, neck formation, zero-point motion and polarization of collective surface vibration as well as static deformation have been identified as key inputs in the understanding of this sub-barrier fusion enhancement [3].For nuclei with a significant static quadrupole deformation [4,5], the main effects are i) on the barrier height (geometrical effect) since it is lower in the elongated direction and ii) on the reorientation of the deformed nucleus (rotational effect) under the torque produced by the long-range Coulomb force. In [6][7][8][9][10], fusion excitation functions were measured for 16 O (spherical) + 144−154 Sm reactions. 144 Sm is spherical whereas 154 Sm is prolate (β 2 ≈ 0.3). The data near the barrier were interpreted as arising from the different orientations of the prolate nucleus. An enhancement of the fusion probability is observed when the deformation axis is parallel to the collision axis ("parallel collision") and a hindrance when the two axis are nearly perpendicular ("perpendicular collision"). In these studies, however, the assumption of an isotropic orientation distribution of the deformed nucleus at contact was made. This contradicts classical calculations [11,12] which show partial reorientation. From the quantum mechanics point of view, the reorientation is a consequence of the excitation of rotational states, which may affect near barrier fusion specially for light deformed nuclei [13,14]. Computational techniques have been developed in the past to solve coupled channel (CC) equations for Coulomb excitation [15-18] but a good understanding of the Coulomb reorientation dynamics during the approach phase is still re...