If that is 40% efficient as in 40% of the theoretical energy input comes out as electricity then it's quite incredible but I find that hard to believe. It would put it in the same range as diesel engines.
The 40% figure is supposed to be "wire-to-wire", but they do list that as the "target efficiency" which suggests it may be somewhat aspirational. It presumably doesn't include the energy needed to extract and refine the oil into whatever kind of burnable fuel you are using, nor the energy necessary to extract and then blend in the sodium additive.
And at the bottom they seem to indicate they are still in the "proving feasibility" stage.
I read this all as: "this is a POC we have, and if we can get it to 40% efficiency than it might make sense (otherwise who cares, just use a conventional generator)"
What does "wire to wire" even mean? The input isn't a wire! (Do they mean they think they can synthesize fuel and burn it at 40% overall efficiency? If so, that's pretty good.)
If you electrolyse water with electricity into h2 and o2 then you have tour first wire.
When you reform the electrons via this engine and the photovoltaic cell you have your second wire.
and better than small diesels / turbines / internal combustion engines, at closer to 20%