It's very nice work those chaps did.
But in the image the molecule does appear to be bent. Probably an artefact of the technique used.
Really? If you consider the pentacene molecule as 'glued' to the surface, with a bit of flexibility in the ring system, then for the molecule to desorb it has to initially detach from the surface at one of the ends and pull up (like a bandaid). If the molecule never fully detaches, but undergoes repeated peeling and sticking at the ends, then those ends will in a time averaged picture appear closer than the centre. Maybe ;-)