In principle, could the time delay between the emission and absorption of a single photon in vacuum be measured?
There are some issues here that I am not sure about in achieving this, namely, uncertainty in the position of the photon at the detecting stage, the production of a single photon on demand and what exactly is meant by a trajectory of one photon in terms of its path length.
Very simplistically, I could envisage a vacuum chamber with a large matrix of single photon detector elements at one end and a single "photon on demand" source at the other. The time between the electronically detected photon and the electronic trigger would then provide a correlation between emission and absorption that could be used to measure the time of flight of a single photon.
I would suppose the uncertainty principle would not allow us to get a detectable hit every time, but given a sufficiently densely packed arrangement of detector elements and if we triggered the source enough times, presumably we would get at least one photon properly striking one detector element and thus allowing us to make the correlation.
As far as the photon on demand is concerned, there seems to be serious research being carried out in achieving this goal, specifically in the field of quantum computing. So presumably such a goal is considered to be a possibility even if it has not yet been achieved. However, is it truly possible to produce one photon on demand with a consistent correlation between the trigger and the actual emission of the photon?