Gravitational potential-based energy storage is nothing new. There are lots and lots of ways this is harnessed; take a grandfather clock, for example. That sai

Gravitational potential-based energy storage is nothing new. There are lots and lots of ways this is harnessed; take a grandfather clock, for example. That said, there are a few points worth noting, if I spend a few minutes thinking about this: * You probably want to maximize "dumb mass", with relation to the lifting mechanism, to bring down your costs this has to be optimized vs density * You probably want to maximize the height difference in order to make the system more effective: --* Less losses due to the "fixed costs", both in time and energy, of positioning the mechanism for the next payload --* If you have lots of blocks to move, or are simply stacking them, it will take up a huge area when discharged, which will at least make the crane system more complicated So, simply stacking blocks does not seem like the perfect solution to me, unless you really want to max out the energy per unit of volume. This could make sense if you were stacking goods that would be useful elsewhere (think a warehouse that doubles as a batteries). Incidentally, batteries, fuel tanks or containers would be a rather good choice if you have to pick one of these. Following on the double-use idea, there is already one distributed, underused infrastructure that is designed for electrically lifting things (though probably not power generation). I can see a few ways that elevators could double as auxiliary power storage units. Of course, special care would be needed to make sure the extra weight doesn't compromises the buildings' structural integrity. The same could be said about virtually any other candidates, which includes existing bridges, dams, cliffs, etc. An interesting take on this would be to see whether it is possible to take it offshore. In theory, given perfect water-proofing (or just good pumps) and logistics, you could change the geometry of a concrete-built floating structure, and lower parts of it extremely deep. Of course, Archimedes doesn't help much with power density, but that wouldn't affect efficiency much (hydrodynamics would, if speed/geometry isn't optimized, though). As a preliminary approach, sea deeps near the coast could make it simpler. I suspect concrete composition would be the hardest part to solve, in that case.